Stem Cells http://www.stemcellsportal.com/
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

First published online June 1, 2006
Stem Cells Vol. 24 No. 9 September 2006, pp. 2085 -2097
doi:10.1634/stemcells.2006-0008; www.StemCells.com
© 2006 AlphaMed Press

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
2006-0008v1
24/9/2085    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Reprints/Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wiese, C.
Right arrow Articles by Wobus, A. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wiese, C.
Right arrow Articles by Wobus, A. M.

TISSUE-SPECIFIC STEM CELLS

Signals from Embryonic Fibroblasts Induce Adult Intestinal Epithelial Cells to Form Nestin-Positive Cells with Proliferation and Multilineage Differentiation Capacity In Vitro

Cornelia Wiesea, Alexandra Rolletscheka, Gabriela Kaniaa, Anne Navarrete-Santosb, Sergey V. Anisimovc, Barbara Steinfarzd, Kirill V. Tarasovc, Sheryl A. Brughc, Ihor Zahaniche, Christiane Rüschenschmidtd,i, Heinz Becki, Przemyslaw Blyszczuka, Jaroslaw Czyza, Jürgen F. Heubache, Ursula Ravense, Olaf Horstmanng, Luc St-Ongeh, Thomas Braunf, Oliver Brüstled, Kenneth R. Bohelerc, Anna M. Wobusa

aIn Vitro Differentiation Group, Institute of Plant Genetics and Crop Plant Research, Gatersleben, Germany;
bInstitute of Anatomy and Cell Biology, University of Halle-Wittenberg, Halle, Germany;
cGerontology Research Center, National Institute on Aging, NIH, Baltimore, Maryland, USA;
dInstitute of Reconstructive Neurobiology, Life & Brain Center, University of Bonn and Hertie Foundation, Bonn, Germany;
eDepartment of Pharmacology and Toxicology, Dresden University of Technology, Dresden, Germany;
fMax Planck Institute for Heart and Lung Research, Bad Nauheim, Germany;
gSurgery University Clinics, Medical Faculty, University of Göttingen, Göttingen, Germany;
hDeveloGen AG, Göttingen, Germany;
iDepartment of Epileptology, University of Bonn, Bonn, Germany

Key Words. Intestinal epithelium • Mouse • Nestin • Embryonic fibroblasts • In vitro differentiation • Wnt • Bone morphogenetic protein

Correspondence: Anna M. Wobus, Ph.D., In Vitro Differentiation Group, Institute of Plant Genetics and Crop Plant Research (IPK), Corrensstrasse 3, D-06466 Gatersleben, Germany. Telephone: +49-39482-5256; Fax: +49-39482-5481; e-mail: wobusam{at}ipkgatersleben.de

Received January 5, 2005; accepted for publication May 16, 2006.

   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 Acknowledgments
 References
 
The intestinal epithelium has one of the greatest regenerative capacities in the body; however, neither stem nor progenitor cells have been successfully cultivated from the intestine. In this study, we applied an "artificial niche" of mouse embryonic fibroblasts to derive multipotent cells from the intestinal epithelium. Cocultivation of adult mouse and human intestinal epithelium with fibroblast feeder cells led to the generation of a novel type of nestin-positive cells (intestinal epithelium-derived nestin-positive cells [INPs]). Transcriptome analyses demonstrated that mouse embryonic fibroblasts expressed relatively high levels of Wnt/bone morphogenetic protein (BMP) transcripts, and the formation of INPs was specifically associated with an increase in Lef1, Wnt4, Wnt5a, and Wnt/BMP-responsive factors, but a decrease of BMP4 transcript abundance. In vitro, INPs showed a high but finite proliferative capacity and readily differentiated into cells expressing neural, pancreatic, and hepatic transcripts and proteins; however, these derivatives did not show functional properties. In vivo, INPs failed to form chimeras following injection into mouse blastocysts but integrated into hippocampal brain slice cultures in situ. We conclude that the use of embryonic fibroblasts seems to reprogram adult intestinal epithelial cells by modulation of Wnt/BMP signaling to a cell type with a more primitive embryonic-like stage of development that has a high degree of flexibility and plasticity.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 Acknowledgments
 References
 
The mammalian intestinal epithelium (IE) is a highly and continuously regenerative tissue whose maintenance and renewal is mediated by intestinal epithelial stem cells. Intestinal stem cells located at the base of the intestinal crypts give rise to transit-amplifying (TA) cells, which proliferate and differentiate into the four specialized cell types of the small intestine: enterocytes, enteroendocrine cells, goblet cells, and Paneth cells [1].

The intestinal stem cell hierarchical pattern of proliferation and differentiation in vivo is well-established [2]. It is thought that stem cells reside within the stem cell compartment or "niche" composed of epithelial and mesenchymal cells and extracellular substrates. The stem cell function is collectively regulated by signals of the niche including signaling molecules of Wnt, hedgehog, bone morphogenetic protein (BMP), and Notch pathways [3–7]. However, due to our limited knowledge about the maintenance of functional stem or progenitor cells by niche factors, to date, researchers have been unable to cultivate multipotent cell lines from the intestine. The immature intestinal epithelial stem cells have not been clearly identified so far, although markers such as the neural stem cell markers Musashi-1 and Hes-1 have been detected in cell populations above the small intestinal crypt [8].

There are, however, two developmental regulatory genes, Cdx1 and Tcf4, which are specifically expressed in the TA cells, that may serve as suitable markers of intestinal epithelial progenitor cells [9, 10]. Specifically, TA cells are characterized by a high but finite capacity for proliferation (dividing ~12–13 hours) and show no clonogenic potential [11].

Because mouse embryonic fibroblasts (MEFs), a mesenchymal cell population, are routinely applied as feeder layers (FLs) to maintain the pluripotency of embryonic stem (ES) cells, we hypothesized that the use of this "artificial niche" might be sufficient to cultivate mouse and human intestinal epithelial cells in vitro, which could be used to identify multipotent Cdx1 and Tcf4-positive cells. Surprisingly, we found that under the influence of signals from FL, intestinal epithelial cells were reprogrammed into cells expressing nestin. Nestin was initially identified as a marker of neural stem and progenitor cells [12], but it has also been detected in many non-neuronal embryonic and adult tissues (i.e., cardiac, skeletal, pancreatic, hepatic), during tissue regeneration [13], and in early committed ES cell derivatives [14].

Using the intermediate filament protein nestin as a marker, we have successfully characterized a novel cell type generated from the intestinal epithelium, which we have termed intestinal epithelium-derived nestin-positive cells (INPs). The derivation of nestin-positive cells from intestinal epithelium by in vitro coculture with FL enabled us to (a) identify a population of Cdx1-positive cells, (b) analyze the properties of nestin-expressing cells with respect to proliferation and to differentiation capacity in vivo and in vitro, and (c) delineate a potential mechanism required for the formation of nestin-positive cells.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 Acknowledgments
 References
 
Cell Culture
Small intestine was prepared from jejunum and ileum of ROSA26, B5/enhanced green fluorescent protein (EGFP), and pNestin-EGFP mice, respectively, and intestinal epithelial cells were isolated by mechanical dissection and enzymatic dissociation with Accutase (PAA Laboratories, Linz, Austria, http://www.paa.at). Cells (n = 5–7 x 107) were cultured according to the scheme shown in Figure 1 on mouse embryonic fibroblast FL [15] in medium I supplemented by leukemia inhibitory factor (LIF), basic fibroblast growth factor (bFGF), and epidermal growth factor (EGF) (supplemental online Methods; supplemental online Table 1). LIF supports the survival and maintenance of embryonic stem and neuronal cells [16], and bFGF and EGF facilitate the proliferation of nestin-positive neural precursor [17, 18] and intestinal epithelial cells [4].


Figure 1
View larger version (63K):
[in this window]
[in a new window]
  		Figure 1. General scheme of the in vitro generation and differentiation of nestin-positive cells from adult mouse and human intestinal epithelium. The cultivation of intestinal epithelium on mouse embryonic fibroblast feeder layers (FL) resulted in small clusters (A) labeled by Hoechst 33342 (B), but without nestin-EGFP-expression (C) of cells isolated from pNestin-EGFP mice at day 5 of primary culture. 1a, Type 1 INPs were generated by continuous subculture (for 5–14 passages) on FL. Shown are type 1 cells (passage 11) grown in the presence of leukemia inhibitory factor (LIF), basic fibroblast growth factor (bFGF), and EGF labeled by ß-galactosidase (D) and nestin (E) [2]. After 9–14 days, compacted clusters appeared that were ß-galactosidase-positive (F) and immunolabeled by nestin (G). Confocal analysis revealed uni- and bipolar neuroblast-like nestin-positive cells (inset). 2a, To generate INPs, 9–14-day-old clusters were plated on gelatin-coated plates and cultured for 1–2 passages for the generation of nestin-positive cells (H). 2b, Neurosphere-like aggregates (J, K) (100 µm in diameter, containing ~100–200 small cells [64]) were generated from cells derived from clusters by culture for 7–14 days [3]. Nine-day-old clusters were isolated by enzymatic dissociation and further cultivated for 5 days in suspension to generate spheroids (L) that were ß-galactosidase-positive (M). Spheroids precultured in the absence of LIF, bFGF, and EGF were smaller and contained fewer nestin-positive cells (40% vs. 60%) [4]. To expand the nestin-positive cell population, spheroids were plated and cultured for 7–14 days on different extracellular matrix proteins. (N): Spheroid-derived cells were ß-galactosidase-positive and expressed nestin (O). (P): Cluster that developed after cultivation of human intestinal epithelium on FL for 9 days. (Q): After transfer of clusters in suspension and culture for 7 days, irregular spheres were formed. (R): Morphology of spheroid-derived cells 12 days after plating of spheroids. (S): Spheroid-derived cells (INPs) expressed nestin. Hoechst 33342 (blue) was used to visualize cell nuclei. Scale bar = 50 µm (A–D, F, J, K–N, P–R), 30 µm (E, G, H, O), or 20 µm (S). Abbreviations: ß-gal, ß-galactosidase; d, days; INP, intestinal epithelium-derived nestin-positive cell; Nest-EGFP, nestin-enhanced green fluorescent protein; PhC, phase contrast.

 


View this table:
[in this window]
[in a new window]
  		Table 1. Transcript abundance of Wnt/BMP signalling factors in feeder cells (feeder layer)

 
For generation of INPs from clusters, at day 9–14, clusters (n = 40 per preparation) were mechanically isolated and cultured in medium I on gelatin-coated culture dishes for an additional 1–2 passages. In parallel, clusters were dissociated by collagenase/dispase and transferred into bacteriological dishes to generate spheroids and neurosphere-like aggregates, respectively. At day 5, spheroids were plated in medium II containing LIF and the cAMP stimulator forskolin (an antiapoptotic factor; [19]) to generate spheroid-derived INPs. In addition, cluster-derived cells were cultured for 7–14 days as neurosphere-like aggregates in media V, Va, and Vb, respectively (supplemental online Table 1). Type 1 cells were generated by continuous subculture on FL in medium I for 5–14 passages (Fig. 1).

Human tissue was obtained from patients by informed consent and after approval by the ethics committee. Tissue of the small intestine (~4-cm pieces) was isolated (after removal of contaminating muscle tissue) and washed in an antibiotic solution, and IE cells were isolated and cultured as clusters and spheroids in a procedure similar to that described for mouse cells (supplemental online Methods).

A detailed description of the mouse models, culture methods, ß-galactosidase staining to identify ROSA26-derived cells, culture of Wnt5a-secreting cells, production of Wnt5a-conditioned medium, and the derivation of clusters and INPs from human intestinal epithelium is given in the supplemental online Methods and Notes.

Clonal Analysis
Spheroid-derived, cluster-derived, and type 1 cells isolated from ROSA26 and B5/EGFP mice were used for the analysis of clonogenic capacity (supplemental online Methods).

Differentiation Analysis
For differentiation into the neuronal lineage, cells were cultured in medium III for 6 days and subsequently in medium IV (all media are given in supplemental online Table 1) for 14 days according to reference 20. Neurosphere-like aggregates (7–14 days in medium V) were plated in medium IV for 14 days and used for reverse transcription-polymerase chain reaction (RT-PCR) analysis. Neuronal differentiation of INPs from clusters (14 days) and of neurosphere-like aggregates (14 days) was tested by coculture on PA6 stromal cells and by incorporation into hippocampal slice cultures (supplemental online Methods). For pancreatic endocrine differentiation [21], cells were cultured in media VI and VIa for 14–21 days, and insulin production was analyzed by enzyme-linked immunosorbent assay (ELISA) (supplemental online Methods). Hepatic differentiation [22] was induced by culture in media VII and VIIa for 14–21 days.

RNA Extraction and RT-PCR Analysis
Total RNA was extracted from MEFs (CD-1 strain), FL (NMRI strain), intestinal epithelial tissue, and type 1 cells after pancreatic and hepatic differentiation and from mouse tissues of embryonic (E)14.5-day embryos. Transcripts from clusters, INPs, and differentiated cells derived from neurosphere-like aggregates, were extracted by Dynalbeads using the Direct micro kit (Dynal Biotech, Oslo, Norway, http://www.dynalbiotech.com). Quantitative-PCR with SybrGreen core reagents (PE Biosystems, Foster City, CA, http://www.appliedbiosystems.com) and RT-PCR were performed as described [15, 23]. Primer sets are provided in supplemental online Table 2.

SAGE Library Construction and Analysis
A SAGE library was constructed from pooled poly A+ RNA (passage 3 MEFs; n = 5) according to Anisimov et al. [23]. Restriction enzymes NlaIII and BsmFI (New England Biolabs, Ipswich, MA, http://www.neb.com) were employed for tag generation. After concatemerization, the SAGE tag containing DNA was cloned into pZeRO-1 and sequenced. Sequencing results were analyzed with SAGE 2000, version 4.12 (http://www.sagenet.org) via GenBank datasets (Unigene Build 132) as previously described [23]. Comparisons among mouse SAGE libraries were performed using publicly available data sets (http://www.ncbi.nlm.nih.gov/geo and http://www.ncbi.nlm.nih.gov/sage) or the Database for Annotation, Visualization and Integrated Discovery (DAVID, http://.david.abcc.ncifcrf.gov) or were identified according to the Bay Area Resource of Mouse Mutations in Secreted and Membrane Proteins (http://socrates.berkeley.edu).

Immunocytochemistry
Immunohistochemistry of intestinal tissue (ROSA26 mice), nestin-EGFP detection in the tissue (pNestin-EGFP mice), and immunocytochemical analysis of FL cells, clusters, type 1 cells, and cells derived from clusters, spheroids, and neurosphere-like aggregates was performed as described [20, 22, 24] (supplemental online Methods; supplemental online Table 3). 5-Bromo-2'-deoxyuridine (BrdU) (10 µmol/l for 12 hours) analysis was done according to the manufacturer’s instructions (Abcam, Cambridge, U.K., http://www.abcam.com).

Electrophysiological Analysis
For patch-clamp analysis, single cells isolated from neurosphere-like aggregates, INPs growing out from isolated clusters, and differentiated (synaptophysin-positive) cells were taken (supplemental online Methods).


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 Acknowledgments
 References
 
Cultivation of Intestinal Epithelial Cells on FL Generated Nestin-Positive Cells (INPs)
We successfully cultivated IE by employing FL (mitomycin [Mutamycin; Bristol-Myers Squibb, Princeton, NJ, http://www.bms.com) C-inactivated MEFs) and media supplemented with LIF, bFGF, and EGF. Because the reprogramming process was dynamic and several cultivation conditions were employed, we have named each morphologically unique cell type according to the characteristics shown in Figure 1. Starting with nestin-negative cells (Fig. 1A–1C), the in vitro conditions rapidly generated nestin-expressing cells, which increased in number with subculture (type 1 cells; Fig. 1E). In the absence of FL, intestinal epithelial cells did not attach, and neither clusters nor INPs formed. Single-cell cultivation did not result in clonal growth as neurospheres. After 1–2 passages, compacted clusters (Fig. 1F) with neuroblast-like, uni- and bipolar nestin-positive cells appeared (Fig. 1G). The number of nestin-positive clusters in the presence of LIF, bFGF, and EGF significantly increased with cultivation time from day 9 (10.1 ± 1.5%) to day 14 (17.2 ± 2.8%; p ≤ .05; Student’s t test) compared with control cultures lacking growth factors (7.1 ± 2.8% and 12.3 ± 4.2%, respectively). Clusters isolated from FL (days 9–14) and replated without FL formed INPs (Fig. 1H). Dissociation of clusters produced aggregates of 20–50 cells that increased in size during cultivation and showed typical neurosphere-like morphology (Fig. 1J), similar to that described in cells from skin [25] and corneal epithelium [26]. Clusters transferred into suspension formed spheroids (Fig. 1L, 1M), which, 2 days after plating, contained up to 60% nestin-positive cells (Fig. 1O).

ß-Galactosidase staining demonstrated that all cluster-derived, spheroid-derived, and nestin-positive cells originated from ROSA26 mice (Fig. 1D, 1F, 1K, 1M, 1N). Early-passage cells demonstrated a normal karyotype (not shown). By applying the same culture methods, we generated nestin-expressing cells from human intestinal epithelium (Fig. 1P–1S; supplemental online Notes).

Origin of INPs
In adult small intestine of pNestin-EGFP mice, we were unable to demonstrate the presence of nestin-EGFP expression in intestinal epithelium (Fig. 2A) and Cdx1-positive transit-amplifying region (Fig. 2B), whereas neurons of the plexus myentericus and submucous plexus were EGFP-positive (Fig. 2A). Nestin-EGFP-positive cells were definitively absent in primary cultures of intestinal epithelial tissue (Fig. 1A–1C, 5 days) and only appeared around the clusters (Fig. 2C) between 8–12 days following coculture with FL cells, indicating that our cultures were not contaminated by enteric neurons.


Figure 2
View larger version (52K):
[in this window]
[in a new window]
  		Figure 2. Origin of INPs. (A): Nestin-EGFP-positive cells are detected only in the plexus myentericus (*) and the submucous plexus (x), but not in the epithelial cells, TA cells, or crypts of the intestine of pNestin-EGFP mice. (B): The transit-amplifying (TA) region of intestinal epithelium of ROSA26 mice is immunostained by Cdx1 (arrowheads). (C): Nestin-EGFP expression appeared in an intestinal epithelial-derived cluster only following cultivation on FL cells for 12 days. The arrowhead marks an EGFP-positive cell with the typical EGFP emission (maximum of emission at 512 nm). (D): Spheroid-derived cells coexpressed nestin (green) and Cdx1 (red). (E): EGFP-positive type 1 cells generated from intestinal epithelium of pNestin-EGFP mice coexpress Cdx1 (red). (F): Nestin-EGFP-positive cells (INPs) derived from clusters do not express Msi-1 (red). Inset shows the immunostaining of Msi-1 in pNestin-EGFP mouse brain-derived cultures used as positive control (cultures according to [65]). (G): Nestin-positive cells (green) from clusters do not coexpress PECAM (red). (H): Nestin-EGFP-positive cells derived from clusters do not express smooth muscle actin (red). (J): Semiquantitative reverse transcription-polymerase chain reaction analysis of mouse intestinal epithelial tissue, clusters at day 9 and INPs derived from clusters. Transcripts were correlated to the housekeeping gene ß-tubulin. (K): Nestin mRNA levels in human intestinal epithelial tissue, clusters (9-day) and INPs were compared with ß-actin transcript levels. Hoechst 33342 (blue) was used to visualize cell nuclei. Scale bar = 20 µm (B, E, H), 30 µm (D), 50 µm (A, C, F, G), 10 µm ([F], inset). Abbreviations: ß-Tub, ß-tubulin; 7-AAD, 7-amino-actinomycin; INP, intestinal epithelium-derived nestin-positive cell; Msi-1, Musashi 1; Nest-EGFP, nestin-enhanced green fluorescent protein; PECAM, platelet/endothelial cell adhesion molecule; sm actin, smooth muscle actin; TPH, tryptophan hydroxylase.

 
By contrast, IE tissue (prior to culture), clusters (9 days) and INPs derived from clusters all showed transcripts to Cdx1 (Fig. 2J). Cdx1 labeling was also observed in the cytoplasm of ~90% of the nestin-positive spheroid-derived (Fig. 2D) and type 1 (Fig. 2E) cells.

Transcripts of Musashi-1 (Msi-1) were only found in the IE tissue, not in clusters or INPs. INPs derived from clusters showed no immunoreactivity to Msi-1 (Fig. 2F), the vascular endothelial cell-specific antigen platelet/endothelial cell adhesion molecule (PECAM) (Fig. 2G), or the subepithelial or pericryptal myofibroblast-specific smooth muscle actin (Fig. 2H). Transcripts of Math1, a marker of intestinal progenitor cells that gives rise to goblet, enteroendocrine, and Paneth cells [27] and tryptophan hydroxylase 1, a marker of serotonin-producing enteroendocrine cells [28], were only found in the IE tissue, not in clusters and INPs. Mash1 transcripts were found in the IE tissue but were absent in clusters and INPs. These data and the absence of p75 in the IE tissue samples (used for culture) argue against a neural crest origin. The data demonstrate that the nestin-positive INPs were not readily detectable in the tissue but rapidly formed under the influence of factors associated with mouse embryonic fibroblast cells.

Characterization of INPs
Nestin-positive cells displayed varying degrees of coexpression with desmin (Fig. 3A–3D), an intermediate filament, which is typical of but not exclusively specific to muscle cells. Approximately 80% of type 1 cells coexpressed nestin and desmin with high BrdU labeling (Fig. 3E, 44.2%), whereas only 15% were nestin+/desmin– (BrdU labeling, 8.3%). Spheroid-derived INPs showed a similar level of BrdU incorporation. Type 1-derived INPs cultured on FL could be propagated for up to 14 passages, whereas spheroid-derived cells cultured without FL showed proliferation for, at most, 5 passages. Exogenous signals supplied by the FL therefore proved critical to support and maintain the proliferation capacity of these cells.


Figure 3
View larger version (43K):
[in this window]
[in a new window]
  		Figure 3. Morphological and functional characterization of INPs. (A–D): INPs derived from spheroids partially coexpress nestin (A, C) and desmin (B, C). (E): The values of nestin+/desmin–/BrdU+ and nestin+/desmin+/BrdU+ cells generated from type 1 and spheroid-derived INPs were determined (in percent) of all cells. The absolute values of nestin+/desmin– and nestin+/desmin+ cells are given in brackets. (F): Morphology of a neurosphere-like aggregate (7 days); the inset shows a dissociated single INP 20 minutes after plating, used for patch-clamp analysis. (G): Outward currents of a representative INP derived from clusters under control conditions, in the presence of 4-AP, and after washing out were elicited from a holding potential of –100 mV. Current-voltage relations are summarized on the right (mean ± SE of n = 7 cells; block by 74 ± 5%; cell capacitance 87 ± 22 pF; p < .05). INPs derived from clusters were also sensitive to 5 mM tetraethylammonium (TEA) (block by 41 ± 6%; n = 7; p < .005; not shown). INPs of neurosphere-like aggregates demonstrated outward currents that were slightly inhibited by TEA (block by 12 ± 4% at +50 mV; n = 4; p < .05, not shown). (H): INPs grown from clusters displayed mainly a bipolar phenotype. (J): Ca2+ and Ba2+ currents in INPs were activated from a holding potential of –95 mV. Original traces are shown for potentials from –75 mV to +5 mV. The current-voltage relations on the right summarize data from nine experiments (cell capacitance, 134 ± 32 pF). (K): Human INPs showed nestin immunoreactivity and partially (L) coexpressed desmin (M). Scale bar = 40 µm (A–D), 50 µm (F, G, K–M), 20 µm ([F], inset). Abbreviations: 4-AP, 4-aminopyridine; BrdU, 5-bromo-2'-deoxyuridine; DIC, differential interference contrast; I, current; INP, nestin-positive cell; Nest, nestin; V, voltage.

 
Patch-clamp experiments performed to characterize functional properties such as electrical excitability and ionic currents revealed the presence of outward currents in some INPs (Fig. 3F, inset) isolated from neurosphere-like aggregates (5 of 25 cells) and in most INPs derived from clusters (18 of 19 cells; Fig. 3H). INPs from clusters possessed K+ currents and low voltage-activated Ca2+ and Ba2+ currents (Fig. 3G, 3J), corresponding, at least partially, to bipolar mouse ES-derived nestin-expressing cells [29]. Na+ currents (required for fast impulse conduction of neurons) were absent in all INPs investigated, suggesting that INPs are immature and do not resemble a neuronal phenotype.

Consistent with cellular properties characteristic of mouse INPs, a high number of human cells also (46.4 ± 21.6%) were colabeled by nestin and desmin (Fig. 3K–3M).

To test the degree of pluripotency of nestin-positive cells, INPs were analyzed for their developmental capacity in vivo and clonogenic capacity in vitro. RT-PCR and histological analysis of embryos developed from INP-transplanted blastocysts indicated that INPs did not participate in tissue formation (supplemental online Fig. 1; supplemental online Notes).

To determine the clonogenic capacity, neurosphere-like aggregates were dissociated, but no secondary aggregates were formed. Also, cultivation of single type 1, cluster- and spheroid-derived INPs in suspension or adherent culture (with different matrix and growth factors; supplemental online Table 1) did not result in the generation of neurosphere-like aggregates or clonal growth, respectively. These observations supported the view that the proliferation of single INPs is either dependent on additional factors or the threshold concentration of these factors is insufficient for clonal growth.

The inability to contribute to tissue formation or to form individual cell clones and the high but finite proliferation capacity of these cells indicate that INPs represent a cell type with properties of progenitor cells. In contrast to primary skin [25] and corneal epithelial cells [26], IE cells in suspension did not form neurospheres, indicating a non-neural cell type.

Signaling Molecules in MEFs/FL Identified by SAGE and Q-PCR Analysis
To identify potential signaling molecules (secreted or cell-cell interacting) implicated in the formation of INPs, we constructed a SAGE library from MEFs (available as part of the Gene Expression Omnibus Database, accession number GSM7759). Library details and tag (transcript) classifications can be found either in supplemental online Table 4 or at http://www.ncbi.nlm.nih.gov/geo.

In silico analysis identified signaling molecules related to Wnt (Wnt1, Wnt2, Wnt5a, Wnt5b, Wnt6, Wnt11, Wnt14, Wisp1, Wisp2, secreted frizzled-related protein [sFRP] 1, sFRP2, Dkk2, and Dkk3) and BMP (BMP1, BMP4, and BMP5) pathways (supplemental online Table 4B). Many of these tags matched multiple transcripts in GenBank, whereas other potential molecules could not be detected because the transcripts either lacked an anchoring enzyme site or the site was immediately adjacent to the poly(A+) sequence. We therefore designed primers (supplemental online Table 2B) for a more systematic Q-PCR analysis of Wnt and BMP signaling molecules. These quantitative analyses confirmed the presence of Wnt- and, to a lesser extent, BMP-associated transcripts in FL (Table 1) and MEFs (supplemental online Table 4D). In FL, sFRP2 and Wnt5a were the most abundant Wnt-associated transcripts, whereas BMP2 and the BMP1 antagonist Ckts1 were the most abundant BMP-associated transcripts.

Figure 4A demonstrates the presence of Wnt4, Wnt5a, Wnt11, sFRP2, sFRP3, BMP2, BMP4, follistatin, and Lef1 transcripts in the mitomycin-C treated embryonic fibroblasts (FL) employed to cultivate IE cells. Wnt3a and Wnt8 transcripts were undetectable, and Wnt1 mRNA was detectable only at low levels. Widespread ß-catenin protein (in the nucleus [Fig. 4B] and cytoplasm and at cell-cell contacts [Fig. 4C]) and the nuclear coexpression of ß-catenin with Lef1 and Tcf4 in nearly all cells (Fig. 4B, 4C) indicate a potentially active canonical Wnt signaling pathway in FL cells. Cellular localization studies show that sFRP2 labeled the cytoplasm of ~90% of FL cells (Fig. 4D), whereas sFRP3 was only rarely detected (Fig. 4E), and Dkk3 showed a punctuate labeling (Fig. 4F).


Figure 4
View larger version (44K):
[in this window]
[in a new window]
  		Figure 4. Reverse transcription-polymerase chain reaction and immunofluorescence analysis of feeder layer (FL) cells. (A): Transcript abundance from genes involved in Wnt/BMP-dependent pathways and Wnt/BMP antagonists relative to the housekeeping gene ß-tubulin. (B–F): Immunofluorescence analysis of FL cells revealed nuclear and cytoplasmic labeling of ß-catenin (B, C) in FL. Lef1 (B) and Tcf4 (C) are abundant in the nucleus. The Wnt-inhibitory sFRP2 (D), sFRP3 (E), and Dkk3 (F) are abundant in FL cells. Scale bar = 20 µm. Abbreviations: BMP, bone morphogenetic protein; Dkk3, Dickkopf 3; sFRP, secreted frizzled-related protein.

 
Transcripts of Wnt/BMP Signaling Components and of Wnt/BMP Target Genes Are Upregulated in INPs
The preceding results suggested that secreted Wnt and BMP signaling molecules from FL might be involved in the transformation of IE cells into nestin-positive INPs. We therefore analyzed the mRNA levels of en-1, emx2, and follistatin, genes that are regulated by Wnt and BMP-dependent signaling pathways. We found that none of these transcripts were detectable in IE tissue, but all three were present in clusters, INPs (Fig. 5A), and spheroid-derived cells (not shown). The greatest abundance of these transcripts was observed in the proliferative INPs. We detected Wnt4 transcripts in IE tissue, but after cultivation of intestinal epithelial cells on FL, these transcripts were upregulated in clusters and INPs. More significantly, Wnt5a and Lef1, which were not detectable in IE tissue, increased in abundance in clusters (Wnt5a) and INPs (Wnt5a, Lef1). Wnt1, Wnt3a, Wnt8, and sFRP2 transcripts were not detected in any of these cell types (not shown). In parallel, transcript levels of BMP2 and BMP4 were found in the IE tissue and in clusters. BMP2 was expressed in INPs at a level comparable to IE tissue, but BMP4 was clearly downregulated in INPs (Fig. 5A).


Figure 5
View larger version (39K):
[in this window]
[in a new window]
  		Figure 5. Reverse transcription-polymerase chain reaction analysis of IE tissue, clusters and INPs from clusters, and immunofluorescence of clusters and INPs. (A): Transcript abundance was analyzed comparatively to ß-tubulin. (B–H): Immunofluorescence was performed on 9-day clusters and INPs derived from clusters for ß-catenin (B, C), Tcf4 (D, E), and Lef1 (F–H), respectively. Secreted frizzled-related protein (sFRP2), SFRP3, and Dkk3 proteins were not detected in clusters or INPs (not shown). Scale bar = 20 µm (B–E, G, H), 10 µm (F). Abbreviations: IE, intestinal epithelium; INP, INP, nestin-positive cell.

 
The cellular distribution and localization of ß-catenin, Tcf4, and Lef1 also changed markedly during the process of transformation of IE cells into INPs. In vivo, stem and TA progenitor cells of the crypts show nuclear localization of ß-catenin [30], whereas in clusters, ß-catenin labeling was low and mainly localized at the membrane (Fig. 5B). In INPs derived from clusters, ß-catenin was highly abundant in the nucleus, cytoplasm, and cellular membranes (Fig. 5C), suggesting a relocation of the protein from the cell membrane to the nucleus (however, a selection for cells that have nuclear ß-catenin may also be possible). In contrast, Tcf4 expressed in TA cells of the IE tissue (not shown) was nearly absent in clusters and completely absent in INPs (Fig. 5D, 5E). Lef1, which is absent in the proliferating compartment of adult intestinal tissue [31], was detected at low levels as intranuclear punctuate structures in the clusters (Fig. 5F) but at relatively high levels in the nuclei of INPs (Fig. 5G, 5H).

INPs Give Rise to Ectodermal and Endodermal Cell Lineages with Differentiation In Vitro
First, we tested INPs for the potential to generate neuronal and glial cells in vitro. We found transcripts for nestin, en-1, synaptophysin, neurofilament, and glial fibrillary acidic protein (GFAP) in cells that differentiate from neurosphere-like aggregates (Fig. 6A; supplemental online Notes). Second, immunofluorescence analysis revealed persistent nestin expression in spheroid- and type 1-derived cells (supplemental online Table 5A). ß-III-Tubulin (Fig. 6B) and GFAP (Fig. 6C) were detected at high levels in spheroid-derived cells.


Figure 6
View larger version (68K):
[in this window]
[in a new window]
  		Figure 6. Differentiation analysis of INPs into neural, pancreatic and hepatic lineages. (A): Reverse transcription-polymerase chain reaction (RT-PCR) analysis of cells derived from neurosphere-like aggregates (7 days) generated from clusters after neuronal differentiation revealed transcripts for nestin, en-1, synaptophysin, NFM, and GFAP. ß-Tubulin is shown as a reference marker. (B, C): Immunofluorescence analysis of spheroid-derived cells differentiated into ß-III-tubulin (B) and GFAP (C)-positive cells. (D, E): Immunocytochemistry of neurosphere-like aggregates, which were cocultured with PA6 stromal cells for 6 days expressing ß-III-tubulin (D) and for 17 days expressing synaptophysin (E). (F): Migration of EGFP-positive INPs from clusters on hippocampal slice cultures (17 days). ß-III-Tubulin-positive (G) and GAP43-EGFP-positive (H) cells observed in hippocampal slice cultures 17 days after transplantation. (J): Plated neurosphere-like aggregates gave rise to neuronal cells with typical neurite extensions and immunoactivity for synaptophysin (inset). During patch-clamp analysis (K), they displayed outward currents that were sensitive to 4-AP. Current-voltage relations of eight cells are shown on the right (cell capacitance, 72 ± 32 pF; p < .05). Voltage protocol is described in the legend of Figure 2. (L–S): After pancreatic differentiation, type 1-derived INPs revealed islet-1 and insulin transcripts by RT-PCR (L). Cells were double-labeled by C-peptide/nestin (M–O), and C-peptide/insulin (P–S). Cells colocalized C-peptide and nestin (27.3%) and islet-1 and nestin (16.5%), but no cytokeratin 19 and nestin colabeling was found. (T–W): After hepatic differentiation, type 1-derived INPs showed HNF3ß, {alpha}-fetoprotein, albumin, and {alpha}1-antitrypsin transcripts (T). Hepatocyte-like cells were labeled by albumin (U), {alpha}1-antitrypsin (V), and cytokeratin 18 (W). Insets show the intracellular localization of proteins at higher magnification. Scale bars = 40 µm, 50 µm (F–H), 20 µm ([J], inset), 10 µm ([U–W], insets), 5 µm ([F], inset). Abbreviations: AAT, {alpha}1-antitrypsin; Alb, albumin; 4-AP, 4-aminopyridine; CA, pyramidal-cell layer; CK, cytokeratin; DG, dentate gyrus; EGFP, enhanced green fluorescent protein; GFAP, glial fibrillary acidic protein; HNF3ß, hepatic nuclear factor 3ß; ß-III Tub, ß-III-tubulin; Nest, nestin; NFM, neurofilament; V, voltage.

 
Cocultivation of EGFP-expressing INPs derived either from clusters (not shown) or from neurosphere-like aggregates on PA6 stromal cells resulted in the differentiation of ß-III-tubulin-expressing cells (Fig. 6D) after 6 days. The majority of cells expressed synaptophysin (Fig. 6E) after 17 days of coculture, but GFAP and tyrosine hydroxylase could not be detected.

After integration into hippocampal slice cultures, all of the transplanted EGFP-positive INPs from clusters (Fig. 6F) and ~50% of INPs from neurosphere-like aggregates (not shown) showed extensive migration into the host tissue. Immunocytochemistry indicated the absence of nestin- and desmin-positive cells, but a strong presence of ß-III-tubulin expression in transplanted INPs (Fig. 6G). GAP43, expressed in axonal growth cones of developing and regenerating neurons and absent in IE-derived clusters on FL, was detected in INPs 17 and 22 days after transplantation (Fig. 6H). GFAP, O4, and peripherin were not detected (absence of other neuronal markers is shown in supplemental online Table 7).

One characteristic property of mature neuronal cells is electrical excitability. Patch-clamp analysis of cells that differentiated from neurosphere-like aggregates (resulting in morphologically typical neuronal [Fig. 6J] and synaptophysin-positive [Fig. 6J, inset] cell types) indicated the presence of outward currents in 29 of 49 cells. The outward currents were inhibited by 63 ± 7% by the K+ channel blocker 4-aminopyridine (4-AP) (Fig. 6K). However, Na+ and Ca2+ currents were still absent, and the cells did not respond to GABA and glycine (not shown), suggesting that the cells have not yet acquired functional properties of mature neuronal cells. INPs differentiating in hippocampal slice cultures for up to 20 days invariably expressed K+ outward currents but no Na+ currents and thus did not acquire mature functions (supplemental online Notes).

Previous studies demonstrated the presence of nestin-positive cells during pancreatic differentiation in vitro [32, 33]. After pancreatic differentiation according to reference 21, type 1 cells revealed transcripts of islet-1 and insulin (Fig. 6L). Immunofluorescence analysis demonstrated the presence of nestin-positive (35%), insulin-positive (51%), and C-peptide-positive (45%) cells, whereas cytokeratin 19 was not detected (supplemental online Table 5B). Double-labeling studies showed that 27.3% of the cells coexpressed nestin and C-peptide (Fig. 6M–6O), indicating that nestin-expressing cells may be involved in the formation of insulin-producing cells. The cells also coexpressed insulin and C-peptide (Fig. 6P–6S). Intracellular insulin levels, analyzed by ELISA, were significantly increased in differentiated type 1 cells; however, the cells did not respond to high glucose concentrations (details are given in supplemental online Table 6), indicating that the cells do not represent a pancreatic endocrine phenotype.

We also analyzed hepatic differentiation of INPs, because nestin-positive cells have been detected during liver regeneration [34] and hepatic differentiation from ES cells [22]. Type 1 cells, which differentiated in the presence of hepatocyte-inducing factors, expressed hepatic nuclear factor 3ß, {alpha}-fetoprotein, albumin, and {alpha}1-antitrypsin transcripts (Fig. 6T). Differentiated type 1 cells were labeled by nestin- (supplemental online Table 5C), albumin- (Fig. 6U), {alpha}1-antitrypsin- (Fig. 6V), and cytokeratin 18- (Fig. 6W) specific antibodies. ELISA did not show albumin secretion (not shown), suggesting that these cells may not have acquired a functional hepatic phenotype.


   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 Acknowledgments
 References
 
Several techniques for the cultivation of intestinal epithelium containing stem and progenitor cells have been described, but the cultivation of these cells invariably led to the loss of differentiation and regeneration potential, probably due to suboptimal culture conditions [35]. It was therefore proposed that epithelial-mesenchymal tissue interactions are important in regulating proliferation and differentiation of intestinal epithelial cells [36]. To mimic this potential interaction, we cocultivated embryonic fibroblasts, which are basically mesenchymal cells that produce signaling factors critical to the in vitro maintenance of ES cells, with intestinal epithelium. Using this method, we identified a previously uncharacterized nestin-positive cell type from mouse and human intestinal epithelium. These results led us to look for potential signaling molecules from FL cells that reprogram intestinal epithelial cells to form INPs. We found that cocultivation of IE cells with FL specifically led to a modification of Wnt/BMP target genes and pathways in INPs.

Origin of Nestin-Positive Cells
Developmentally, it was important to determine the origin of these novel cells. The expression of Cdx1 in the nestin-positive cells suggests that INPs originate from the proliferative TA region of the intestinal epithelial crypt and not from interstitial cells of Cajal, which are Cdx1-negative mesenchymal cell derivatives [9, 37] that often express smooth muscle actin (but it is also possible that INPs originate from another Cdx1-negative cell type).

The absence of Math1 suggests that INPs are not derivatives of intestinal progenitors, and the lack of tryptophan hydroxylase 1 expression indicates that INPs are not derived from serotonin-producing enteroendocrine cells. The lack of smooth muscle actin expression in INPs also argues against the subepithelial or pericryptal myofibroblast nature of INPs. Because INPs showed no immunoactivity for PECAM, we exclude an origin from vascular endothelial cells. However, we cannot rule out the possibility that stem cells found at the base of the crypt are induced to generate the Msi-1-negative/nestin-positive cells when cultivated in the presence of MEF-FL cells.

In addition, there is no evidence for a neural crest origin [38], because of the absence of p75 in intestinal tissue, and mash1 and Wnt1 [39] in the clusters and INPs [40]. The lack of neurosphere formation from primary intestinal epithelial cells and the inability to generate secondary neurospheres from cluster-derived neurospheres furthermore suggests a non-neural precursor cell type. We therefore propose that the most likely origin for INPs is the TA region of the intestinal epithelial crypt.

The use of cell isolates from transgenic mice also demonstrated that all the IE cell descendants (clusters, spheroids, INPs, and differentiated cell types) described in this study showed ß-galactosidase staining and originated from ROSA26 mice and not from contaminating feeder cells. The absence of nestin-positive cells in adult intestinal epithelium of pNestin-EGFP mice also confirmed that nestin-expressing cells (INPs) developed only following coculture with FL cells.

Potential Mechanisms Involved in the Generation of INPs
The comparative SAGE analyses implicated Wnt signaling molecules as abundant factors in FL. Because Wnt signaling molecules are known to influence cell lineage decisions [41] and because of the role of Tcf4 in maintaining the IE stem cell compartment [10], we studied transcript and protein abundance of members of the Wnt/BMP signaling cascade. We show that Wnt4 is highly abundant in FL cells and in IE-derived clusters cultured on FL, but because Wnt4 is maximally expressed in clusters and not INPs, various signaling molecules or concentrations of molecules (supplied by FL cells and medium components) may be involved in its regulation. Because high Wnt5a transcript levels were found in the SAGE catalogue from feeder cells and because INPs revealed upregulated Wnt5a levels, the presence of Wnt5a mRNA in clusters and especially in INPs could implicate the activation of a noncanonical Wnt signaling pathway [42]. Wnt5a is highly expressed in embryonic limb bud mesenchyme and ectoderm [43], and in the same regions, nestin-positive cells have been found [44]. Because Wnt5a is regulated by bFGF or BMPs in vivo and in vitro [45], we hypothesize that similar mechanisms may be implicated in INP formation.

During embryogenesis, Cdx1 is localized in the cytoplasm and nuclei of cells in the prospective crypts at E15.5, but at E17.5 days post conception, Cdx1 is generally restricted to the nucleus [45]. The accumulation of Lef1 and loss of Tcf4 (a direct regulator of Cdx1 [46]) may account for the cytoplasmic distribution of Cdx1 expression observed in INPs. This finding and the downregulation of Tcf4 suggest a modulation of Wnt/ß-catenin signaling pathways in the process of reprogramming of intestinal epithelial cells into INPs.

The upregulation of en-1, emx2, and follistatin, direct targets of Wnt and BMP signaling, suggests a direct role of these factors in the generation of INPs [47–49]. Follistatin antagonizes BMP4 activity during neural differentiation [50], and its upregulation, concomitant with the downregulation of BMP4 in nestin-positive INPs, further suggests the implication of BMP signaling in INP formation. The preponderance of Wnt signaling molecules in FL and the activation of Wnt/BMP target genes in nestin-positive INPs, therefore, led us to hypothesize that INPs form, at least partially, through the interactions of Wnt and BMP signaling molecules released by FL cells. Still, when we cultured intestinal tissue in Wnt5a-conditioned medium (supplemental online Notes), no changes in the abundance of nestin-positive clusters and in the nestin transcript level were observed. This leads us to speculate that perhaps more than one signaling pathway may be involved in INP formation and further experiments are necessary to confirm the relationship between INP formation and Wnt and BMP activation.

In summary, the reprogramming process of adult intestinal epithelial cells by feeder cells resulted in the generation of INPs that possess some properties of embryonic intestinal epithelium. The cells show cytoplasmic Cdx1 expression and Lef1 (instead of Tcf4) signaling comparable to IE cells during embryogenesis [31]. The use of an artificial niche of MEFs therefore seems to reprogram adult tissue cells to a cell type with a more primitive embryonic stage of development that has a high degree of flexibility and plasticity.

Properties of Nestin-Positive Cells and Derivatives In Vivo and In Vitro
Three lines of evidence indicate that INPs represent a cell type expressing properties of a progenitor and not a stem cell type. First, after injection into blastocysts, INPs did not contribute to tissue formation indicating an insufficient developmental capacity in vivo. In contrast, in vitro INPs demonstrate a high degree of plasticity, and under appropriate conditions, INPs differentiate into cells with properties typical of ectoderms (neuronal and glial) and endoderms (pancreatic and hepatic). Second, the nestin-positive cells demonstrate a high but finite proliferation capacity, a trait that is characteristic of progenitor cells [51]. Type 1 cells incorporate BrdU, similar to that of nestin-positive cells derived from R1 ES cells [13] and from limbal-corneal epithelium [26]. However, after continued cultivation (5 passages for clusters and spheroids, 14 passages for type 1 cells), the in vitro proliferation capacity decreases significantly. Third, we were unable to clonally isolate and cultivate single nestin-positive cells. Altogether, these properties suggest that INPs represent a cell type that expresses some characteristics of progenitor cells.

In contrast to ES-derived specialized cells, the functional capacity of INPs seemed to be limited. Our observation that INP derivatives of synaptophysin-positive cells were devoid of Na+ currents and ligand-gated ion channels (typical of neuronal cells), and the lack of functional properties of hepatic and pancreatic cells suggests that the in vitro conditions do not fully support the development of functional cell types or that the intrinsic capacity of INPs is not able to accomplish functional competence. Importantly, transplantation into hippocampal slice cultures showed extensive INP migration, paralleled by a decrease of nestin and desmin and differentiation into ß-III-tubulin- and GAP43-positive cells. However, the limited functional (electrophysiological) maturation suggests a restricted neuronal fate specification of INPs, potentially due to a lack of appropriate niche signals. Alternatively, selection by culture conditions, especially the effects of bFGF and EGF [52], might account for this limitation because similar findings were obtained with limbal epithelial-derived neuron-like cells [26].

Similar nestin-positive cell types characterized by restricted differentiation capacity and lack of neuron-specific ion channel activity have been isolated from various somatic tissues, including bone marrow and cord blood [53–55], mouse skin dermis [25], inner ear [56], or human corneal epithelium [57]. However, our finding that nestin-positive cells differentiate into cells expressing some pancreatic cell markers are supported by recent findings that human pancreas-derived nestin-positive cells develop into islet-like cells after long-term cultivation [58].

Nestin Expression as a Marker of Cells Representing High Plasticity
Our results and data from other publications [21, 22, 32] suggest that nestin might be a general marker useful for the identification of a diverse potential progenitor cell type. In adult organisms, nestin is restricted to distinct locations associated with regenerative processes [13], but in response to injury, nestin is detected in myoblasts [59], and it is re-expressed in astrocytes [60]. The coexpression of desmin with nestin in INPs and in differentiating ES cells [13] may also be consistent with a cell type of high plasticity. This is supported by the isolation of INP derivatives characterized by properties typical for neuronal, hepatic, and pancreatic cells. Normally, in adult cells, nestin and desmin do not form heteropolymers, but a copolymerization of nestin and desmin has been described for skeletal muscle cells of E15.5 days p.c. embryos [61]. Nestin and desmin coexpression was also found in liver cell types under regenerative conditions [34].

Nestin was also detected in gastrointestinal stromal cell tumors [37]. Because INPs showed a temporary proliferation capacity and euploid karyotypes, we exclude the possibility that nestin expression in INPs is correlated with tumorigenic cell transformation. These results, and the observation that the duodenum, jejunum, and ileum are largely cancer-free [62], argue against a tumorigenic status of INPs.

The molecular mechanisms for nestin upregulation (or reactivation) after injury are not understood. One possibility could be that soluble factors released by the adjacent tissue (niche cells) trigger gene activation, similar to that for growth factors in the brain [63] and as suggested for BMP4 in carious lesions of teeth [64]. The present study implicates Wnt/BMP signaling in the induction of INPs, and we hypothesize that similar mechanisms may be involved in processes of in vivo tissue regeneration. Because cellular responses after injury recapitulate processes of early development, we suggest that in adult somatic tissues, nestin expression/re-expression in response to injury and/or secreted factors of niche cells may represent a status associated with flexibility and plasticity of cells.


   DISCLOSURES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
Acknowledgments
 References
 
The authors indicate no potential conflicts of interest.


   ACKNOWLEDGMENTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 Acknowledgments
References
 
We are grateful to Drs. M. Shamblott (Johns Hopkins University School of Medicine, Baltimore, MD), H.-H. Arnold (Technical University of Braunschweig, Braunschweig, Germany), and Neil Theise (Department of Medicine, Beth Israel Medical Center, New York, NY) for helpful discussions. We thank Drs. B. Meyer (Max Planck Institute for Biophysical Chemistry, Göttingen, Germany) for Cdx1 antibodies, T. Brand (Technical University of Braunschweig), and A. Kispert (Hanover Medical School, Hanover, Germany) for supplying the 3T3 and Wnt5a-expressing cell lines and G. Kempermann (Max-Delbrück-Center, Berlin) for pNestin-EGFP mice. We thank S. Sommerfeld, O. Weiss, and K. Meier for excellent technical assistance. This work was supported by the University Clinics Grosshadern of the Ludwigs-Maximilians University Munich (institutional grant by Prof. B. Reichardt) and the Aventis Foundation (C.W.); by the Intramural Research Program of the NIH, National Institute on Aging (K.R.B.); and by the Deutsche Forschungsgemeinschaft (WO 503/3-2), the Bundesministerium für Bildung und Forschung (01GN0106), DeveloGen AG, Göttingen, and Fonds der Chemischen Industrie, Germany (A.M.W.). C.W. and A.R. contributed equally to this work. Current addresses for C.W.: Department of Anatomy and Embryology, Academic Medical Center, University of Amsterdam, The Netherlands; for G.K. and P.B.: Experimental Critical Care Medicine Group, University Hospital Basel, Switzerland; for J.H.: Schering AG, Berlin; for J.C.: Faculty of Biotechnology, University Krakow, Poland; for L.S.-O.: Affectis Pharmaceuticals, Munich.


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 Acknowledgments
 References
 

  1. Leedham SJ, Brittan M, McDonald SA. Intestinal stem cells. J Cell Mol Med 2005;9:11–24.[Medline]

  2. Potten CS, Loeffler M. Stem cells: Attributes, cycles, spirals, pitfalls and uncertainties. Lessons for and from the crypt. Development 1990;110:1001–1020.

  3. Winton DJ. Stem cells in the epithelium of the small intestine and colon. In: Marshak DR, Gardner RL, Gottlieb D, eds. Stem Cell Biology. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 2001;515–536.

  4. Dignass AU, Sturm A. Peptide growth factors in the intestine. Eur J Gastroenterol Hepatol 2001;13:763–770.[CrossRef][Medline]

  5. Clatworthy JP, Subramanian V. Stem cells and the regulation of proliferation, differentiation and patterning in the intestinal epithelium: Emerging insights from gene expression patterns, transgenic and gene ablation studies. Mech Dev 2001;101:3–9.[CrossRef][Medline]

  6. He XC, Zhang J, Tong WG. BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling. Nat Genet 2004;36:1117–1121.[CrossRef][Medline]

  7. Fre S, Huyghe M, Mourikis P. Notch signals control the fate of immature progenitor cells in the intestine. Nature 2005;435:964–968.[CrossRef][Medline]

  8. Kayahara T, Sawada M, Takaishi S et al. Candidate markers for stem and early progenitor cells, Musashi-1 and Hes1, are expressed in crypt base columnar cells of mouse small intestine. FEBS Lett 2003;535:131–135.[CrossRef][Medline]

  9. Subramanian T, Marchionini D, Potter EM. Striatal xenotransplantation of human retinal pigment epithelial cells attached to microcarriers in hemiparkinsonian rats ameliorates behavioral deficits without provoking a host immune response. Cell Transplant 2002;11:207–214.[Medline]

  10. Korinek V, Barker N, Moerer P. Depletion of epithelial stem-cell compartments in the small intestine of mice lacking Tcf-4. Nat Genet 1998;19:379–383.[CrossRef][Medline]

  11. Bach SP, Renehan AG, Potten CS. Stem cells: The intestinal stem cell as a paradigm. Carcinogenesis 2000;21:469–476.[Abstract/Free Full Text]

  12. Lendahl U, Zimmerman LB, McKay RD. CNS stem cells express a new class of intermediate filament protein. Cell 1990;60:585–595.[CrossRef][Medline]

  13. Wiese C, Rolletschek A, Kania G et al. Nestin expression: A property of multi-lineage progenitor cells? Cell Mol Life Sci 2004;61:2510–2522.[CrossRef][Medline]

  14. Kania G, Corbeil D, Fuchs J. Somatic stem cell marker prominin-1/CD133 is expressed in embryonic stem cell-derived progenitors. STEM CELLS 2005;23:791–804.[Abstract/Free Full Text]

  15. Wobus AM, Guan K, Yang HT. Embryonic stem cells as a model to study cardiac, skeletal muscle, and vascular smooth muscle cell differentiation. Methods Mol Biol 2002;185:127–156.[Medline]

  16. Li M, Sendtner M, Smith A. Essential function of LIF receptor in motor neurons. Nature 1995;378:724–727.[CrossRef][Medline]

  17. Gritti A, Cova L, Parati EA. Basic fibroblast growth factor supports the proliferation of epidermal growth factor-generated neuronal precursor cells of the adult mouse CNS. Neurosci Lett 1995;185:151–154.[CrossRef][Medline]

  18. Okabe S, Forsberg-Nilsson K, Spiro AC. Development of neuronal precursor cells and functional postmitotic neurons from embryonic stem cells in vitro. Mech Dev 1996;59:89–102.[CrossRef][Medline]

  19. Branton RL, Love RM, Clarke DJ. cAMP included during cell suspension preparation improves survival of dopaminergic neurons in vitro. Neuroreport 1998;9:3223–3227.[Medline]

  20. Rolletschek A, Chang H, Guan K et al. Differentiation of embryonic stem cell-derived dopaminergic neurons is enhanced by survival-promoting factors. Mech Dev 2001;105:93–104.[CrossRef][Medline]

  21. Blyszczuk P, Czyz J, Kania G. Expression of Pax4 in embryonic stem cells promotes differentiation of nestin-positive progenitor and insulin-producing cells. Proc Natl Acad Sci U S A 2003;100:998–1003.[Abstract/Free Full Text]

  22. Kania G, Blyszczuk P, Czyz J. Differentiation of mouse embryonic stem cells into pancreatic and hepatic cells. Methods Enzymol 2003;365:287–303.[CrossRef][Medline]

  23. Anisimov SV, Tarasov KV, Tweedie D et al. SAGE identification of gene transcripts with profiles unique to pluripotent mouse R1 embryonic stem cells. Genomics 2002;79:169–176.[CrossRef][Medline]

  24. Meyer BI, Gruss P. Mouse Cdx-1 expression during gastrulation. Development 1993;117:191–203.[Abstract]

  25. Toma JG, Akhavan M, Fernandes KJ. Isolation of multipotent adult stem cells from the dermis of mammalian skin. Nat Cell Biol 2001;3:778–784.[CrossRef][Medline]

  26. Zhao X, Das AV, Thoreson WB. Adult corneal limbal epithelium: A model for studying neural potential of non-neural stem cells/progenitors. Dev Biol 2002;250:317–331.[CrossRef][Medline]

  27. Yang Q, Bermingham NA, Finegold MJ. Requirement of Math1 for secretory cell lineage commitment in the mouse intestine. Science 2001;294:2155–2158.[Abstract/Free Full Text]

  28. Legay C, Faudon M, Hery F. 5-Ht Metabolism in the intestinal wall of the rat. 1. The Mucosa. Neurochem Int 1983;5:721–727.[CrossRef]

  29. Lenka N, Lu ZJ, Sasse P. Quantitation and functional characterization of neural cells derived from ES cells using nestin enhancer-mediated targeting in vitro. J Cell Sci 2002;115:1471–1485.[Abstract/Free Full Text]

  30. Batlle E, Henderson JT, Beghtel H. Beta-catenin and TCF mediate cell positioning in the intestinal epithelium by controlling the expression of EphB/ephrinB. Cell 2002;111:251–263.[CrossRef][Medline]

  31. Wong MH, Huelsken J, Birchmeier W. Selection of multipotent stem cells during morphogenesis of small intestinal crypts of Lieberkuhn is perturbed by stimulation of Lef-1/beta-catenin signaling. J Biol Chem 2002;277:15843–15850.[Abstract/Free Full Text]

  32. Zulewski H, Abraham EJ, Gerlach MJ. Multipotential nestin-positive stem cells isolated from adult pancreatic islets differentiate ex vivo into pancreatic endocrine, exocrine, and hepatic phenotypes. Diabetes 2001;50:521–533.

  33. Seaberg RM, Smukler SR, Kieffer TJ. Clonal identification of multipotent precursors from adult mouse pancreas that generate neural and pancreatic lineages. Nat Biotechnol 2004;22:1115–1124.[CrossRef][Medline]

  34. Niki T, Pekny M, Hellemans K. Class VI intermediate filament protein nestin is induced during activation of rat hepatic stellate cells. Hepatology 1999;29:520–527.[CrossRef][Medline]

  35. Evans GS, Flint N, Somers AS. The development of a method for the preparation of a rat intestinal epithelial cell primary cultures. J Cell Sci 1992;101:219–231.[Abstract/Free Full Text]

  36. Fukamachi H. Proliferation and differentiation of fetal rat intestinal epithelial cells in primary serum-free culture. J Cell Sci 1992;103:511–519.[Abstract]

  37. Tsujimura T, Makiishi-Shimobayashi C, Lundkvist J. Expression of the intermediate filament nestin in gastrointestinal stromal tumors and interstitial cells of Cajal. Am J Pathol 2001;158:817–823.[Abstract/Free Full Text]

  38. Kruger GM, Mosher JT, Bixby S. Neural crest stem cells persist in the adult gut but undergo changes in self-renewal, neuronal subtype potential, and factor responsiveness. Neuron 2002;35:657–669.[CrossRef][Medline]

  39. Brault V, Moore R, Kutsch S. Inactivation of the beta-catenin gene by Wnt1-Cre-mediated deletion results in dramatic brain malformation and failure of craniofacial development. Development 2001;128:1253–1264.[Abstract]

  40. Lo L, Sommer L, Anderson DJ. MASH1 maintains competence for BMP2-induced neuronal differentiation in post-migratory neural crest cells. Curr Biol 1997;7:440–450.[CrossRef][Medline]

  41. Kleber M, Sommer L. Wnt signaling and the regulation of stem cell function. Curr Opin Cell Biol 2004;16:681–687.[CrossRef][Medline]

  42. Jönsson M, Smith K, Harris AL. Regulation of Wnt5a expression in human mammary cells by protein kinase C activity and the cytoskeleton. Br J Cancer 1998;78:430–438.[Medline]

  43. Parr BA, Shea MJ, Vassileva G. Mouse Wnt genes exhibit discrete domains of expression in the early embryonic CNS and limb buds. Development 1993;119:247–261.[Abstract]

  44. Wroblewski J, Engstrom M, Edwall-Arvidsson C. Distribution of nestin in the developing mouse limb bud in vivo and in micro-mass cultures of cells isolated from limb buds. Differentiation 1997;61:151–159.[CrossRef][Medline]

  45. Chimal-Monroy J, Montero JA, Ganan Y. Comparative analysis of the expression and regulation of Wnt5a, Fz4, and Frzb1 during digit formation and in micromass cultures. Dev Dyn 2002;224:314–320.[CrossRef][Medline]

  46. Lickert H, Domon C, Huls G. Wnt/(beta)-catenin signaling regulates the expression of the homeobox gene Cdx1 in embryonic intestine. Development 2000;127:3805–3813.[Abstract]

  47. Pizette S, Abate-Shen C, Niswander L. BMP controls proximodistal outgrowth, via induction of the apical ectodermal ridge, and dorsoventral patterning in the vertebrate limb. Development 2001;128:4463–4474.[Medline]

  48. Bartholin L, Maguer-Satta V, Hayette S. Transcription activation of FLRG and follistatin by activin A, through Smad proteins, participates in a negative feedback loop to modulate activin A function. Oncogene 2002;21:2227–2235.[CrossRef][Medline]

  49. Theil T, Aydin S, Koch S. Wnt and Bmp signalling cooperatively regulate graded Emx2 expression in the dorsal telencephalon. Development 2002;129:3045–3054.[Abstract/Free Full Text]

  50. Fainsod A, Deissler K, Yelin R. The dorsalizing and neural inducing gene follistatin is an antagonist of BMP-4. Mech Dev 1997;63:39–50.[CrossRef][Medline]

  51. Seaberg RM, van der Kooy D. Adult rodent neurogenic regions: The ventricular subependyma contains neural stem cells, but the dentate gyrus contains restricted progenitors. J Neurosci 2002;22:1784–1793.[Abstract/Free Full Text]

  52. Hack MA, Sugimori M, Lundberg C. Regionalization and fate specification in neurospheres: The role of Olig2 and Pax6. Mol Cell Neurosci 2004;25:664–678.[CrossRef][Medline]

  53. Kabos P, Ehtesham M, Kabosova A et al. Generation of neural progenitor cells from whole adult bone marrow. Exp Neurol 2002;178:288–293.[CrossRef][Medline]

  54. Wislet-Gendebien S, Leprince P, Moonen G. Regulation of neural markers nestin and GFAP expression by cultivated bone marrow str