Factors Regulating Stem Cell Biology in Development and DiseaseView this Special Issue
Comparative Microarray Analysis of Proliferating and Differentiating Murine ENS Progenitor Cells
Postnatal neural progenitor cells of the enteric nervous system are a potential source for future cell replacement therapies of developmental dysplasia like Hirschsprung’s disease. However, little is known about the molecular mechanisms driving the homeostasis and differentiation of this cell pool. In this work, we conducted Affymetrix GeneChip experiments to identify differences in gene regulation between proliferation and early differentiation of enteric neural progenitors from neonatal mice. We detected a total of 1333 regulated genes that were linked to different groups of cellular mechanisms involved in cell cycle, apoptosis, neural proliferation, and differentiation. As expected, we found an augmented inhibition in the gene expression of cell cycle progression as well as an enhanced mRNA expression of neuronal and glial differentiation markers. We further found a marked inactivation of the canonical Wnt pathway after the induction of cellular differentiation. Taken together, these data demonstrate the various molecular mechanisms taking place during the proliferation and early differentiation of enteric neural progenitor cells.
The enteric nervous system (ENS) is a largely autonomous and highly complex neuronal network found in the gastrointestinal tract. Its two major plexuses are integrated into the layered anatomy of the gut wall and, together with central modulating influences, exert control over gastrointestinal motility, secretion, ion-homeostasis, and immunological mechanisms . In order to achieve this variety of functions, the ENS is composed of a multitude of different neuronal and glial cell types and closely interacts with smooth muscle cells and myogenic pacemaker cells called interstitial cells of Cajal. Furthermore, a population of neural stem or progenitor cells in the ENS has been identified in rodents [2, 3] and humans that retain their proliferative capacity throughout adult life even into old age [4, 5]. It is therefore not surprising that the correct functioning of the ENS as well as the regulation on enteric neural progenitor cells is subjected to the influence of a myriad of transmitters, neurotrophic and growth factors, signalling molecules, and extracellular matrix components, which are not exclusively expressed by neural cell types . Likewise, the control of the development of the ENS is equally complex and mutations in its genetic program can lead to fatal dysplasia like Hirschsprung’s disease (HCSR) [7, 8].
HSCR is hallmarked by an aganglionic distal bowel leading to life-threatening disturbances in intestinal motility. Today’s therapeutic gold standard, the surgical resection of the affected gut segments, is nevertheless associated with problematic long-term outcomes with regard to continence . In order to improve the therapeutic success, the use of autologous enteric neural stem cells was proposed . This concept relies on the in vitro expansion of enteric neural stem cells derived from small biopsy materials. However, we are just beginning to understand the molecular mechanisms that underlie neural stem cell biology and how this knowledge can be used for optimizing in vitro culture conditions [11, 12].
Genome-wide gene-expression analyses are a useful tool to examine the genetic programs and cellular interactions and have been widely used to identify potential markers or signalling mechanisms especially in CNS neurospheres or cancer tissues. Further, gene-expression assays have also helped to unravel genetic prepositions associated with HSCR [13, 14], though little effort has so far been put into characterizing the genetic profile of enteric neural stem cells in vitro .
Here, we used an Affymetrix microarray analysis to evaluate the genetic expression profile of proliferating murine enteric neural stem cells and its changes during the early differentiation in vitro.
2. Materials and Methods
2.1. Cell Culturing
Cell culturing was conducted as described previously . The handling of animals was in accordance to the institutional guidelines of the University of Tuebingen, which conform to the international guidelines.
Neonatal (P0) C57BL/6 mice without regard to sex were decapitated and the whole gut was removed. After removal of adherent mesentery the longitudinal and circular muscle layers containing myenteric plexus could be stripped as a whole from the small intestine. Tissue was chopped and incubated in collagenase type XI (750 U/mL; Sigma-Aldrich, Taufkirchen, Germany) and dispase II (250 μg/mL; Roche Diagnostics, Mannheim, Germany) dissolved in Hanks’ balanced salt solution with Ca2+/Mg2+ (HBSS; PAA, Pasching, Austria) for 30 min at 37°C. During enzymatic dissociation the tissue was carefully triturated every 10 min with a fire polished 1 mL pipette tip. Prior to the first trituration step, cell suspension was treated with 0.05% (w/v) DNAse I (Sigma-Aldrich). After 30 min, tissue dissociation was stopped by adding fetal calf serum (FCS; PAA) to a final concentration of 10% (v/v) to the medium. Undigested larger tissue pieces were removed with a 40 μm cell strainer (BD Biosciences, Franklin Lakes, NJ, USA). Residual enzymes were removed during two washing steps in HBSS at 200 g. After dissociation, cells were resuspended in proliferation culture medium (Dulbecco’s modified Eagle’s medium with Ham’s F12 medium (DMEM/F12; 1 : 1; PAA)) containing N2 supplement (1 : 100; Invitrogen, Darmstadt, Germany), penicillin (100 U/mL; PAA), streptomycin (100 μg/mL; PAA), L-glutamine (2 mM; PAA), epidermal growth factor (EGF; 20 ng/mL; Sigma-Aldrich), and fibroblast growth factor (FGF; 20 ng/mL; Sigma-Aldrich). Cells were seeded into 6-well plates (BD Biosciences) in a concentration of 2.5 × 104 cells/cm2. Only once before seeding, the medium was supplemented with B27 (1 : 50; Invitrogen). EGF and FGF were added daily and culture medium was exchanged every 3 days. All cultivation steps were conducted in a humidified incubator at 37°C and 5% CO2. An overview of the following cell culture protocol is shown in Figure 1. During proliferation phase of the culture, cells formed spheroid-like bodies termed enterospheres. After 5 days of proliferation, free-floating enterospheres were picked and transferred to petri dishes (Ø 60 mm; Greiner Bio One, Frickenhausen, Germany) in 5 mL fresh proliferation medium and proliferation was continued for further 4 days.
Single free-floating enterospheres (50 enterospheres/dish) were picked again, washed 3 times in Tris buffer, and transferred into new petri dishes containing either proliferation medium or differentiation medium. Differentiation medium consists of DMEM/F12 containing N2 supplement (1 : 100), penicillin (100 U/mL), streptomycin (100 μg/mL), L-glutamine (2 mM), and ascorbic acid-2-phosphate (200 μM; Sigma-Aldrich).
Enterospheres were proliferated or differentiated for 2 more days, thereby forming the two experimental groups “proliferation” and “differentiation.” The difference in expression between those two groups (differentiation versus proliferation) was successively compared by microarray analysis as described below.
2.2. Affymetrix Microarray Analysis
Affymetrix microarray analysis was conducted similar to previously published data in three independent experiments, each with cell cultures prepared from 2 pups from the same litter . In each experiment, free-floating enterospheres were picked as described above in order to diminish the fraction of adhesive fibroblasts and smooth muscle cells.
Total RNA of enterospheres of both groups was extracted using the RNeasy Micro Kit (Qiagen). RNA quality was evaluated on Agilent 2100 Bioanalyzer with RNA integrity numbers (RIN) of the samples in this study being in the range from 8 to 10. RIN numbers higher than 8 are considered optimal for downstream application .
Double-stranded cDNA was synthesized from 100 ng of total RNA, subsequently linearly amplified, and biotinylated using the GeneChip WT cDNA Synthesis and Amplification Kit (Affymetrix, Santa Clara, CA, USA) according to the manufacturer’s instructions. 15 μg of labeled and fragmented cDNA was hybridized to GeneChip Mouse Gene 1.0 ST arrays (Affymetrix). After hybridization, the arrays were stained and washed in a Fluidics Station 450 (Affymetrix) with the recommended washing procedure. Biotinylated cDNA bound to target molecules was detected with streptavidin-coupled phycoerythrin, biotinylated anti-streptavidin IgG antibodies and again streptavidin-coupled phycoerythrin according to the protocol. Arrays were scanned using the GCS3000 GeneChip Scanner (Affymetrix) and AGCC 3.0 software. Scanned images were subjected to visual inspection to check for hybridization artifacts and proper grid alignment and analyzed with Expression Console 1.0 (Affymetrix) to generate report files for quality control.
Normalization of raw data was performed by the Partek Software 6.6, applying an RMA (Robust Multichip Average) algorithm. Significance was calculated using a -test without multiple testing correction (Partek), selecting all transcripts with a minimum change in expression level of 1.5-fold together with a value less than 0.05.
In this study, we investigated the changes of the genetic expression profile that occur during the transition from proliferating to differentiating enteric neural progenitor cells in vitro. Therefore, we generated enterospheres by 9 day in vitro cultures, which then could be picked and either proliferated or differentiated for two more days (Figure 1). mRNA was subsequently extracted and gene expression of these two groups was analysed by Affymetrix microarray analysis.
Analysis of mRNA expression was performed on a GeneChip Mouse Gene 1.0 ST array that determines the expression profile of 28.853 genes. Each gene was interrogated by a median of 27 probes that are spread along the full gene.
In total, the gene chip detected 1454 transcripts to be at least 1.5-fold differentially expressed between proliferating and differentiating enterospheres. 1333 of these transcripts code for already identified proteins. 541 genes were found to be upregulated and 792 genes were found to be downregulated in comparison to proliferating enterospheres (see Supplementary Table of the Supplementary Material available online at http://dx.doi.org/10.1155/2016/9695827).
We used the ingenuity pathway analysis software (IPA) and data mining with the science literature search engine http://www.ncbi.nlm.nih.gov/pubmed/ to divide the genes into different groups according to their function during cellular development. The largest functional group contained 171 genes related to cell cycle and apoptosis (Table 1, Supplementary Table ). Here, we identified especially different cyclin proteins and cell division cycle proteins that were mainly downregulated. Further, we found several genes that are linked to neural development as well as genes regulating neural stem cell proliferation and differentiation. Furthermore, we also detected neuronal and glial differentiation markers and numerous genes involved in synapse formation (Table 2). It is noteworthy that we also identified a group of genes that are known to be involved in the differentiation of smooth muscle cells (Table 3) as well as in extracellular matrix components (Table 4). Additionally, we found regulated genes related to canonical Wnt signalling indicating a deactivation of this pathway during ENS progenitor cell differentiation (Figure 2, Table 5).
The proliferation and differentiation of enteric neural progenitor cells during embryonic and postnatal development are controlled by a complex interplay of various intrinsic and extrinsic factors. Their exact timing is crucial for proper migration and proliferation of neural crest cells and for their differentiation into the various neural cell types that compose the complex neural structures of the ENS. Although research in recent years extended our understanding of ENS development and its pathologies , there are still many genes and processes unknown. Particularly, factors regulating neural progenitor proliferation and differentiation in the developing and postnatal gut as well as cellular and molecular interaction systems remain largely elusive. Here, we used in vitro cultures of enteric neural progenitor cells derived from murine tunica muscularis to scan for molecular programs and signalling pathways acting on cell proliferation and early differentiation.
Our experiment aimed to elucidate gene regulations in enterospheres that occur while ENS progenitor cells leave their proliferative state and begin to differentiate into more defined and specific cell types. The results of the Affymetrix gene expression analysis showed the up- and downregulation of overall 1333 known genes that code for already identified proteins. 171 of these genes could be linked to cell proliferation (Table 1, Supplementary Table ). Amongst them we detected genes coding for proteins related to the kinetochore complex (like NSL1 , NUF2 , SKA1-3 , and ZWILCH ), cyclin proteins , cyclin-dependent kinases (CDK) , and several types of centromere proteins. The regulation of 145 of these genes strongly indicates a slowdown of cell cycle progression as it was intended by the experimental deprivation of growth factor supplementation by the end of the proliferation phase (see Section 3). Interestingly, betacellulin (BTC) was upregulated nearly 6-fold although it was reported to promote cellular proliferation in the neural stem cell niche . Nonetheless, the vast majority of genes including all regulated cyclins, cell division cycle proteins, and kinetochore proteins were found to be downregulated.
We also checked the regulated genes for apoptosis markers to see whether the stop in proliferation was related to cell death (Supplementary Table ). Since only 3 of 12 apoptotic genes were regulated in the direction that indicates apoptosis, it is unlikely that apoptosis played a leading role in the interruption of proliferation. Still, the effect and regulation of apoptosis during enteric sphere cultures are an important cornerstone of understanding enteric neural progenitors in culture and in vivo and require further investigation. Together, on a broad basis, this dataset provides strong evidence that this cell culture design is applicable to decreasing the proliferative rate of enteric neural progenitor cells without inducing cell death or apoptosis in an appreciable quantity.
To further evaluate the proliferative conditions of cell types present in enterospheres, we focused on different cell specific markers of neural progenitors as well as neurons, glial, or smooth muscle cells. We consider this complex cellular composition of the enterospheres an advantage compared to more purified neural crest derived neurospheres as we are able to capture complex interactions and secretion mechanisms between cell types that might also play an important role in vivo. Interestingly, we found 8 genes involved in adult central or embryonic neural stem cells homeostasis (Table 2). The majority of genes like EPHA2  are regulated in a way that suggests that neural stem cells exit the proliferative cell cycle to enter differentiation programs. This idea was supported by the upregulation of numerous genes that drive neuronal and glial differentiation like NEUROD4  or OLIG1 . In this context, we identified several upregulated genes involved in proper myelination. As enteric and central glia cells are known to temporally express myelin-related proteins during development, it is conceivable that this regulation is part of the early glial differentiation program . Moreover, also typical markers of differentiated neurons (class III beta-tubulin, CALB2 ) and enteric glia (GFAP ) were found to be upregulated. Intriguingly, S100B, a common glia cell marker, was downregulated contrasting the rest of our data. Again, this might be due to the complex differentiation program of enteric glia, in which S100B plays a role at later stages.
Furthermore, the establishment of neuronal cell communication was strongly regulated. Here, we found an increased expression of genes related to synaptogenesis (LRRTM2 and 3 , neurotrimin ) and to SNARE or vesicle protein function (STXBP3, SV2C , and SYT6 ). We also identified a number of genes involved in transmitter metabolism (COMT, DDC) as well as neurotransmitter receptor like 5-HT, glutamate, and adrenergic receptors. However, the regulation of those genes was highly variable shedding light on the intricacy of synapse formation in the developing enteric nervous system. This complexity is carried on by genes related to axon sprouting and guidance like semaphorins  or RGMa .
Additionally, we found that regulated genes directly involved in the differentiation of muscle cells and/or enteric pacemaker cells called interstitial cells of Cajal (Table 3). Particularly interesting is the upregulation of a number of genes known to drive smooth muscle differentiation like ARID5B , FOSL2  and genes that are expressed in differentiated smooth muscle cells in the intestine like AFAP1 , ENPP2 , and CNN1  as well as various myosin and actin isoforms. These data confirm the fact that cultured spheroids are composed of different cell types present in the intestinal tunica muscularis and further indicate that deprivation of growth factors induces differentiation of smooth muscle cells resembling molecular processes in the developing gut. In fact, we among others were previously able to confirm the presence of smooth muscle cells derived from enterosphere culture by BrdU-immunolabeling costudies . However, it is noteworthy that a few genes related to muscular differentiation (endoglin , smoothelin , NUP210 , caldesmon 1 , and ACTN1 ) were downregulated contrasting the expression pattern observed in the majority of regulated genes. This hints to complex regulatory mechanisms controlling the myogenic differentiation program in which these genes are not required at all or in a different temporal sequence not mapped by our experimental design. It is further remarkable that five markers expressed in interstitial cells of Cajal (ICC) including KIT  were downregulated.
Moreover, the regulation of 43 extracellular matrix proteins like collagens, integrins, proteoglycans, and matrix metallopeptidases points to a reconstruction of extracellular environment that has been discussed to influence neural stem cell behaviour  (Table 4). Taken together, these results illustrate the ongoing genetic programs during early differentiation of enterospheres.
Within the dataset, it was of special interest to find particularly many regulated genes related to the canonical Wnt pathway (Table 5). The involvement of canonical Wnt signalling has frequently been shown in the regulation of various stem cell niches, like intestinal epithelium or CNS derived neural stem cells. However, these studies exhibited different and partly contradicting outcomes, which strongly hint to the variable functions of canonical Wnt signals in different tissues during embryonic and postnatal development. In previous work, we found regulation of several Wnt-related genes in the context of thyroid hormone dependent differentiation of enteric neural progenitor cells indicating a potential role of the canonical Wnt pathway activation during the proliferation of this progenitor cell pool . Canonical Wnt signalling has frequently been reviewed in the literature—just recently by Ring et al. . In brief, secreted Wnt proteins bind to frizzled receptors (FZD) complexed with low density lipoprotein receptor-related protein 5/6 (LRP5/6) coreceptors. Thereafter, the scaffolding protein disheveled (DVL) is recruited to FZD and inhibits the β-catenin destruction complex (AXIN2, APC, and GSK-3β). Therefore, β-catenin accumulates in the cytoplasm and translocates to the nucleus where it binds to TCF/LEF transcription factors to initiate Wnt target gene expression. Interestingly, our current data strongly indicate that the canonical Wnt pathway is switched off during the first two days of enteric progenitor differentiation on several levels of the signalling cascade (Figure 2). On the one hand we identified a downregulation of activating parts of the signalling cascade itself like the receptor proteins FZD7 and LRP5 or the transcription factors TCF19, TCF7L1, and LEF1. On the other hand, inactivating elements of the pathway like parts of the β-catenin destruction complex AXIN2 and LRRK2  were upregulated. We also found numerous modulators of the signalling cascade. It is of interest that the majority of those genes are reported to inhibit the signalling process extracellularly or on receptor level (Notum , FRZB , DKK2 , and LRP4 ), in the cytoplasm (NEDD4L , NKD1 , PRICKLE1 , NOV , and APOE ), or in the nucleus (TLE3 , EDIL3 ). Furthermore, we identified target genes of the canonical Wnt pathway that were either upregulated (e.g., AXIN2 that exerts a negative feedback on the pathway) or downregulated like the cell cycle progression genes CCND1 and SPRY4 . We also found a lower expression of SPRY2 , a Wnt target gene and known inhibitor of GDNF signalling , in the differentiation group. Together with a strong upregulation of GDNF itself by 4.325-fold, this might drive enteric progenitor cells into neural differentiation .
Taken together, it is conceivable that canonical Wnt signalling plays a role in the maintenance of the enteric progenitor pool during proliferation and is switched off at the beginning of differentiation conditions. Indeed, our previous gene expression analyses  as well as recently published cell culture experiments  and yet unpublished in vitro analyses strongly support this hypothesis.
This study focused on the changes in gene expression of enteric neural progenitor cells occurring within the first two days of transition from a proliferative state to differentiation in vitro. Using microarray analysis, we found a marked inhibition of cell cycle progression in general as well as strong evidence for neural stem cells differentiation into enteric neurons and glia cells. These findings were substantiated by the upregulation of genes related to synapse formation and neural connectivity. Most interesting, we found that this transition from enteric neural progenitor proliferation to differentiation was accompanied by a considerable inactivation of the canonical Wnt signalling pathway. This, together with previous work, strongly indicates that canonical Wnt activation is one of the driving mechanisms of enteric neural progenitor proliferation and thus might play a role in the homeostasis of this cell pool in vivo and in vitro.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Peter Helmut Neckel and Roland Mohr contributed equally to this work.
The project was supported by a grant from the German Federal Ministry for Education and Research (01GN0967). The authors would like to thank Andrea Wizenmann, Andreas Mack, and Sven Poths for their helpful advice.
The supplementary material contains tables with additional information about the gene expression analysis.
D. Natarajan, M. Grigoriou, C. V. Marcos-Gutierrez, C. Atkins, and V. Pachnis, “Multipotential progenitors of the mammalian enteric nervous system capable of colonising aganglionic bowel in organ culture,” Development, vol. 126, no. 1, pp. 157–168, 1999.View at: Google Scholar
G. M. Kruger, J. T. Mosher, S. Bixby, N. Joseph, T. Iwashita, and S. J. Morrison, “Neural crest stem cells persist in the adult gut but undergo changes in self-renewal, neuronal subtype potential, and factor responsiveness,” Neuron, vol. 35, no. 4, pp. 657–669, 2002.View at: Publisher Site | Google Scholar
T. A. Heanue and V. Pachnis, “Expression profiling the developing mammalian enteric nervous system identifies marker and candidate Hirschsprung disease genes,” Proceedings of the National Academy of Sciences of the United States of America, vol. 103, no. 18, pp. 6919–6924, 2006.View at: Publisher Site | Google Scholar
M. V. Gómez-Gaviro, C. E. Scott, A. K. Sesay et al., “Betacellulin promotes cell proliferation in the neural stem cell niche and stimulates neurogenesis,” Proceedings of the National Academy of Sciences of the United States of America, vol. 109, no. 4, pp. 1317–1322, 2012.View at: Publisher Site | Google Scholar
M. Kolatsi-Joannou, X. Z. Li, T. Suda, H. T. Yuan, and A. S. Woolf, “In early development of the rat mRNA for the major myelin protein P0 is expressed in nonsensory areas of the embryonic inner ear, notochord, enteric nervous system, and olfactory ensheathing cells,” Developmental Dynamics, vol. 222, no. 1, pp. 40–51, 2001.View at: Publisher Site | Google Scholar
M. Metzger, S. Conrad, T. Skutella, and L. Just, “RGMa inhibits neurite outgrowth of neuronal progenitors from murine enteric nervous system via the neogenin receptor in vitro,” Journal of Neurochemistry, vol. 103, no. 6, pp. 2665–2678, 2007.View at: Google Scholar
L. E. Peri, K. M. Sanders, and V. N. Mutafova-Yambolieva, “Differential expression of genes related to purinergic signaling in smooth muscle cells, PDGFRα-positive cells, and interstitial cells of Cajal in the murine colon,” Neurogastroenterology & Motility, vol. 25, no. 9, pp. e609–e620, 2013.View at: Publisher Site | Google Scholar
Y. Ding, Y. Zhang, C. Xu, Q.-H. Tao, and Y.-G. Chen, “HECT domain-containing E3 ubiquitin ligase NEDD4L negatively regulates Wnt signaling by targeting dishevelled for proteasomal degradation,” The Journal of Biological Chemistry, vol. 288, no. 12, pp. 8289–8298, 2013.View at: Publisher Site | Google Scholar
K. Sakamoto, S. Yamaguchi, R. Ando et al., “The nephroblastoma overexpressed gene (NOV/ccn3) protein associates with Notch1 extracellular domain and inhibits myoblast differentiation via Notch signaling pathway,” The Journal of Biological Chemistry, vol. 277, no. 33, pp. 29399–29405, 2002.View at: Publisher Site | Google Scholar
A. Takai, H. Inomata, A. Arakawa, R. Yakura, M. Matsuo-Takasaki, and Y. Sasai, “Anterior neural development requires Del1, a matrix-associated protein that attenuates canonical Wnt signaling via the Ror2 pathway,” Development, vol. 137, no. 19, pp. 3293–3302, 2010.View at: Publisher Site | Google Scholar
Y. Katoh and M. Katoh, “FGF signaling inhibitor, SPRY4, is evolutionarily conserved target of WNT signaling pathway in progenitor cells,” International Journal of Molecular Medicine, vol. 17, no. 3, pp. 529–532, 2006.View at: Google Scholar