SEARCH

SEARCH BY CITATION

Keywords:

  • apolipoprotein B;
  • enteric nervous system;
  • Hirschsprung’s disease;
  • Neuro2a cell line;
  • RetC620R

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contribution
  11. References
  12. Supporting Information

Background  GDNF/RET and Endothelin-3 (ET-3)/EDNRB regulate survival, differentiation, migration, and proliferation of neural crest-derived cells. Although several RET and EDNRB signalling mediators have been characterized, most of the genes targeted by these two pathways are still largely unknown. We focused our study on apolipoprotein B (APOB) as a novel target gene of the RET and EDNRB pathways, based on previous data obtained using a Caenorhabditis elegans strain mutant for the homologue of mammalian ECE1.

Methods  Molecular and cellular studies of Apob were performed in the murine Neuro2a cells, an in vitro model for studying neural crest-derived cell development, along with a mouse knock-in for the Hirschsprung-associated mutation RetC620R. Silencing for Apob and Ret has been performed via shRNA.

Key Results  GDNF/RET and ET-3/EDNRB cooperated in inducing neuronal differentiation resulting in Apob activation in Neuro2a cell line. Apob expression was downregulated in mouse embryos homozygous for the RetC620R mutation and presenting a severe Hirschsprung phenotype. Ret silencing prevented Apob expression increase. MAPK P38 kinase activation evoked Apob expression via GDNF/RET signalling in Neuro2a cells. A p53-dependent repressor element in Apob promoter resulted in a reduced Apob expression. Silencing of Apob reduced HuD protein expression.

Conclusions & Inferences  Apob is a novel downstream target of the RET/EDNRB pathways with a role in neuronal survival and maintenance, as indicated by its effect on HuD expression. Our data provide a conceptual framework to investigate and establish the role of APOB gene in severe gut dysmotility.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contribution
  11. References
  12. Supporting Information

An intact and functioning enteric nervous system (ENS) is required for life. The segmental absence of the ENS, which occurs either as a congenital [Hirschsprung’s disease (HSCR)] or acquired defect,1 causes a severe impairment of gastrointestinal transit. Hirschsprung’s disease is characterized by the absence of enteric ganglia in the intestinal tract, with an incidence of approximately 1/5000 births affecting more men than women (4 : 1 ratio).2,3

RET is the main gene involved in HSCR: the majority of patients harbor either heterozygous mutations in the coding region or a heterozygous or homozygous hypomorphic allele in a conserved enhancer sequence in intron 1.4,5 Mutations in other genes encoding proteins involved in either RET or Endothelin receptor type-B (EDNRB) signalling pathways are found in a small percentage of HSCR patients.2,3

RET codes for a tyrosine kinase transmembrane receptor, with two alternative isoforms at the C-terminal recruiting different signalling factors.6 Activation of RET occurs upon self-dimerization, induced by the binding of glial cell line-derived neurotrophic factor (GDNF) family ligand to glycosylphosphatidylinositol-anchored co-receptor GDNF family receptor α and resulting in the activation of three main kinases, Akt, ERK, and MAPK P38.4 RET activation regulates ENS normal development by modulating survival, differentiation, migration, and proliferation of enteric neural crest-derived cells.

EDNRB is a G-coupled receptor that binds any of three endothelins (ET-1, -2, -3),7 produced as preproendothelins and proteolytically activated by Endothelin-converting Enzyme-1 (ECE-1). The crosstalk between GDNF/RET and ET-3/EDNRB pathways in coordinating ENS development has been demonstrated in in vitro and in vivo studies.8–11 Nevertheless, the molecular bases of this crosstalk remain to be elucidated.

This study was designed to investigate apolipoprotein B (APOB) as a novel target of the crossregulation between RET and EDNRB pathways, as emerged from previous studies using the Caenorhabditis elegans strains mutant for the mammalian ECE1 homologue.12APOB is expressed in two isoforms, APOB100 and APOB48, the former being synthesized in the liver as an essential component of lipoproteins, and the latter being generated by RNA editing in the gut for chylomicron production.13APOB mutations cause familial hypobetalipoproteinemia14 and familial defective apolipoprotein B-100.15

A correlation between APOB and the genes involved in ENS development, such as EDNRB and RET, has not been reported so far. However, Apob knockout mice showed severe neural tube defects suggesting a role for APOB during embryogenesis and in particular with regard to neural crest development.16 We therefore focused on the Neuro2a murine cell line, an in vitro model for studying neural crest development and neuronal differentiation,17,18 along with a mouse knock-in for the mutation RetC620R associated to HSCR19 to investigate the role of Ret and Ednrb signalling pathways in Apob expression and to identify the molecular partners involved in this process. We demonstrated that Apob expression is actually activated by GDNF/Ret via MAPK P38 and modulated by ET-3/Ednrb. Furthermore, we identified p53 as a transcription factor involved in the regulation of Apob expression. Taken together our data provide a basis to investigate the role of APOB in the clinical setting specifically focusing on severe gut dysmotility disorders.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contribution
  11. References
  12. Supporting Information

Cell cultures and growth factor treatments

Murine Neuro2a cells were grown in DMEM, supplemented with 10% fetal bovine serum (FBS), 2 mm l-glutamine, at 37 °C in a 5% CO2 humidified atmosphere. SK–N–MC and SK–N–BE were grown in DMEM/F12 1/1 v/v and RPMI, respectively, supplemented with 2 mm l-glutamine, 100 U mL−1 penicillin, 100 μg mL−1 streptomycin and 10% FBS at 37 °C in a 5% CO2 humidified atmosphere.

After 16 h serum starvation, Neuro2a cells were treated with 50 ng mL−1 GDNF (R&D Systems, Minneapolis, MN, USA) and 300 nmol L−1 ET-3 (Sigma-Aldrich, St. Louis, MO, USA), separated and in combination for different times. To test RET-activated signalling pathways, Neuro2a cells were shifted to a serum free medium in the presence of kinase inhibitors before GDNF treatment, as follows: TCN (Triciribine V) (Akt inhibitor) 10 μm 24 h; SB202190 (MAPK38 inhibitor) 20 μm 24 h; PD98059 (ERK inhibitor) 40 μm 24 h. To test Neuro2a cell growth and differentiation upon GDNF and/or ET-3 treatment, Neuro2a cells were seeded on PolyD lysine (Sigma-Aldrich) precoated 24-well plates to a final density of 2500 or 5000 cells. The following day the medium was replaced with DMEM supplemented with 2 mm l-glutamine, 100 U mL−1 penicillin, 100 μg mL−1 streptomycin, 0.5 μg mL−1 DMPH4, 2.5 μg mL−1 L-GSH, 50 μg mL−1 L-ascorbic acid, and N2 Supplement (synthetic medium) and cells were treated with GDNF (10 ng mL−1) and ET-3 (300 nmol L−1) separately and in combination. Triplicate wells were studied for each condition, including three wells in which cells were grown in the synthetic medium without the addition of any growth factors (control cultures). For time course observations, cells were plated in multiple 24-wells plates, fixed in methanol (−20 °C) at a particular time point (every 24 h) for up to 5 days. Half of the medium was replaced once every 2 days. At the end of the treatment, cell growth was evaluated by counting cells present in each well using phase-contrast microscopy with the help of a grid placed in the ocular and lines drawn on a transparent film attached to the 24-multiwell plate in such a way as to obtain distinct fields to be evaluated separately. Cell differentiation was evaluated by a detailed analysis of cell morphology by phase-contrast microscopy. Cells were classified according to the presence and shape of neurites in six groups: undifferentiated, round-shaped cells; cells with one neurite (monopolar), cells with two opposite neurites (bipolar); cells with two opposite and branched neurites (branched-bipolar), cells with many neurites (star cells); cells with one prevalent branched neurite (branched) (Figure S1).

Mice knock-in RetC620R

Ret C620R mice were grown at Charles River animal facility (Lecco, Italy). Experimental procedures were approved by the ethical committee for animal experimentation of the University of Bologna, in accordance with European legislation (European Communities Council directive of November 24, 1986, 86/609/EEC).

Western blotting

Whole-cell lysates were obtained by scraping cells directly in IP Buffer 1X containing deoxycholate (Sigma-Aldrich) and a cocktail of protease inhibitors (Roche Diagnostics, Indianapolis, IN, USA) and phosphatase inhibitors (Sigma-Aldrich) on ice for 15 min. Cells were collected and lysed through an insulin syringe. Protein concentration was evaluated with the DC protein concentration assay Kit (Biorad, Hercules, CA, USA). Proteins were denatured by heating at 95 °C for 5 min in Laemli buffer, separated by SDS gel electrophoresis, transferred onto nitrocellulose membrane (GE Healthcare, Chalfont St. Giles, UK), and subjected to Western blotting. Bands were visualized using the ECL method (GE Healthcare).

Primary antibodies used were as follows: anti-PAN-RET 1 : 200, anti-phospho-Akt 1 : 1000, anti-total Akt 1 : 1000, anti-phospho-Erk 1 : 1.000, anti-total Erk 1 : 1000 (Santa Cruz, Biotechnology Inc., Santa Cruz, CA, USA), anti-phospho MAPK P38 kinase 1 : 1000, anti-MAPK P38 kinase 1 : 1000 (Cell Signaling, Danvers, MA, USA), anti-HuD 1 : 100 (Invitrogen, Carlsbad, CA, USA), anti-γ-tubulin 1 : 10 000, anti-vinculin 1 : 30 000 (Sigma-Aldrich).

Ret and Apob silencing via shRNA

Neuro2a cells (5 × 105) were transfected either with the pRS-shRNA vectors (OriGene, Rockville, MD, USA) (specific for mouse Ret, TR1504590, and Apob silencing, TI562711), or the empty vector, or the scramble vector (control) using lipofectamine (Invitrogen). After 24 h, cells were divided into the number of wells necessary for GDNF stimulation. The next day, cells were shifted to a serum-free medium for 16 h. GDNF treatment and protein extraction were performed as previously described. Lack of Ret expression was verified via Western blotting using the PAN-RET antibody (1 : 200). RNA was extracted and Apob silencing was verified by real-time quantitative RT-PCR as described below.

Immunocytochemisty

Neuro2a cells were fixed in 4% PFA, washed in PBS and incubated in a 0.3% H2O2/Methanol solution for 15 min, then incubated for 60 min at 4 °C in blocking solution (1% BSA/PBS) and washed three times for 30 min in PBS. Two primary anti-RET antibodies were used obtaining similar results: anti-Ret antibody (R&D Systems) and anti-PAN-RET (Santa Cruz). Cells were incubated with primary antibody at a 1 : 200 dilution for 2 h, washed in PBS and incubated for 1 h with biotinylated secondary antibody (Santa Cruz) at a 1 : 200 dilution. Slides were washed three times for 30 min in PBS. Staining was performed using the Vectastain Elite ABC kit (Vector, Burlingame, CA, USA) and the DAB substrate kit for peroxidase (Vector). SK–N–MC and SK–N–BE neuroblastoma cell lines were used as negative and positive control, respectively.20,21

Immunohistochemistry

Immunohistochemistry was performed on paraffin-embedded adult human colon tissues (patients routinely operated for uncomplicated colon cancers commonly used as controls for immunohistochemical experiments) according to protocols validated in our laboratory.22 Briefly, tissue sections were treated to remove paraffin embedding by three sequential washes in xylene and graded ethanol. Antigen unmasking was performed by heating sections at 95 °C in 10 mm Sodium citrate buffer, pH 6.0, for 10 min and subsequent cooling at room temperature for 30 min. To reduce endogenous peroxidases, tissue sections were treated with an ad hoc blocking kit (GeneTex Inc., Irvine, CA, USA) and then incubated for 16 h at 4 °C with primary antibodies, i.e. anti-human APOB diluted 1 : 100 (ABCAM, Cambridge, UK); anti-human NSE diluted 1 : 250 (Santa Cruz). Fluorescent secondary antibodies were diluted in Triton-X (1 : 300) and incubated for 60 min at room temperature, before subsequent dehydration and visualization under a LEICA DMLB fluorescent microscope.

Quantitative real-time polymerase chain reaction (qRT-PCR) analysis

Total RNA was extracted using the Trizol method and was reverse transcribed using Superscript III kit and random hexamers (Invitrogen). Primers for Apob, Actb were designed with Primer3 (http://www.genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi) and RTPrimerDB (http://medgen.ugent.be/rtprimerdb/). Quantitative real-time polymerase chain reaction was performed with SYBRGreen, 0.5 μm primers, in an ABI 7500 Real-Time PCR System (Applied Biosystems, Foster city, CA, USA). All target genes were normalized with the corresponding endogenous control using the ΔCt comparative method; Apob expression levels were normalized to beta-actin (Actb).

Expression of Ret, Gfra1 and Ednrb in Neuro2a cells was assessed by RT-PCR using mouse brain RNA as positive control. All primers used in the present work are available on request.

Expression of Apob in mouse embryos RetC620R

Ret C620R /Ret C620R , +/RetC620R and wild-type embryos were removed by Caesarian section from pregnant carrier mothers at 16 days post coitum (d.p.c.) and genotyped as described previously.19 Tissues were preserved in RNAlater (QIAGEN, Valencia, CA, USA) and total RNA was extracted using the Trizol method. qRT-PCR was performed as described above.

Mouse Apob promoter analysis

Mouse Apob upstream region was amplified using primer 5′-GGGCTCGAGGGCTTATGGCCTCTGTATGC-3′ inserting an XhoI restriction site at the 5′ end, and primer 5′-GGGAAGCTTCCTGCTGCTTGCTCTACAGG-3′ inserting a HindIII restriction site at the 3′ at the following condition: 60 ng mouse genomic DNA, 0.4 μm primers, 0.4 mm dNTPs, 2.5 mm MgCl2, DMSO 5%, 0.2 μL Platinum Taq in a final volume of 50 μL. The PCR amplification was performed as follows: 10 min 94 °C, 1 min at 94 °C, 30 s at 62 °C, 1 min at 72 °C, 46 cycles. The PCR product was inserted in the pGL3 basic vector and different deleted clones were generated with the Erase-a-base kit (Promega, Madison, WI, USA). The mutated promoter construct at −986 bp was generated with the QuickChange site-directed mutagenesis kit (Agilent, Santa Clara, CA, USA). 400 ng of each construct was independently co-transfected with 400 ng of the control vector containing the renilla luciferase reporter into 2 × 105 Neuro2a cells using lipofectamine in duplicate. 24 h after transfection, cells were shifted to a serum free medium for 16 h, treated for 1 h with GNDF (50 ng mL−1) and lysed in PLB 1X (Passive Lysis Buffer) for the Dual-Luciferase Reporters assay (Promega). Firefly luciferase values were normalized to renilla luciferase values.

Electromobility shift assay (EMSA) on mouse Apob -986bp region

107 Neuro2a cells were rinsed in cold PBS-Na3VO4 0.2 mm, pelleted by centrifugation at 1600 g and lysed in 400 μL of Buffer A (10 mm HEPES pH 7.9, 10 mm KCl, 0.1 mm EDTA, 1 mm DTT, 1 mm Na3VO4 and a protease inhibitors cocktail) for 15 min in ice. A quantity of 25 μL of 10% NP-40 was added and the samples were centrifuged at 16 000 g 30 s. The nuclear pellets were resuspended in 50 μL Buffer C (20 mm HEPES pH 7.9, 0.4 mm KCl, 0.1 mm EDTA, 0.1 mm EGTA, 1 mm DTT, 1 mm Na3VO4 and a protease inhibitors cocktail), incubated on ice for 45 min, and centrifuged at 16 000 g 10 min at 4 °C. Supernatants were subjected to EMSA or stored at −80 °C until use. EMSA shift was performed with the LightShift Chemiluminescent EMSA Kit (Thermo Scientific Inc., Pittsburg, PA, USA). A quantity of 10 μg of nuclear extracts, 20 fmol of 5′ biotin-ATGCATTTCTATGCGTGCATGCCCATGAA probe, 50 ng of poly(dI), 2.5% glycerol, 0.05% NP-40 and 5 mm MgCl2, were incubated at room temperature for 20 min in 20 μL. For the supershift assay, nuclear extracts were pre-incubated with anti-p53 mouse monoclonal antibody (0.6 μg; Santa Cruz) on ice for 20 min prior to probe addition. For competition assay, 100-fold excess unlabeled double-stranded oligonucleotides were incubated. Bound complexes were separated on 4% polyacrylamide gels, blotted onto nylon membrane and visualized according to the LightShift Chemiluminescent EMSA Kit.

Statistical analysis

Statistical analysis was performed using Statgraphics Plus 5.1. software (Statistical Graphics Corp., Warrenton, VA, USA). The Student’s t-test was used to compare two groups, anova and chi-square test were used for multiple comparisons. A value of P < 0.05 was considered significant. Data are expressed as means ± SE.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contribution
  11. References
  12. Supporting Information

ET-3/Ednrb signalling modulates GDNF/Ret-related cell growth and differentiation and Apob expression

We tested the existence of a crosstalk between Ednrb and Ret in Neuro2a cells and established whether or not these molecular pathways modulated Apob expression. The expression of Ret, Gdnfra, and Ednrb in Neuro2a cells has been verified by RT-PCR, whereas Ret protein has been detected using immunocytochemistry (Fig. 1A and B).

image

Figure 1.  Effect of GDNF and ET-3 treatment on Neuro2a cell number and differentiation. (A) Expression of Ret, Gfra1, and Ednrb in mouse brain (1) and Neuro2a cells (2); (3) negative PCR controls. (B) Ret protein expression in Neuro2a cells. Immunolabeling with anti-Ret primary antibody of SK-N-MC that do not express Ret (1) and Neuro2a cell lines (3). (2, 4) Negative controls of NK-N-MC and Neuro2a cells, respectively. (C) Neuro2a cell numbers in cultures plated at moderate-high cell-plating densities at the end of a 4 days treatment with GDNF (10 ng mL−1) and ET-3 (300 nm) separated and in combination. Data represent means ± SE (D) Total differentiated Neuro2a cells in cultures plated at low cell-plating densities (2500 cells per well) at the end of a 4 days treatment with GDNF (10 ng mL−1) and ET-3 (300 nmol L−1) separated and in combination. (E) Cell-phenotype distribution in the analyzed cells. Data represent percentage of differentiated cells (±SE) observed in samples of 500 cells per replica-well for treatment conditions. Cells were classified in one of the six different cell phenotypes: monopolar, bipolar, branched-bipolar, star, and branched cells.

Download figure to PowerPoint

At a density of 5000 cells per well, the administration of GDNF exerted a more potent stimulatory effect on Neuro2a cell proliferation compared to ET-3 (anova, P < 0.05; chi-squared test, P < 0.0001). As expected, the concomitant administration of both ligands resulted in a lower Neuro2a cell proliferation due to molecular interference of the two pathways (Fig. 1C). At a density of 2500 cells per well, Neuro2a cells treated with GDNF and ET-3 for 4 days showed the most evident differentiation effect. Differentiated Neuro2a cells were classified into six different phenotypes (Figure S1). Differentiated cells were observed as a result of Neuro2a treatment with either GDNF or ET-3 (chi-squared test, P = 8.04 × e−05 and P = 0.04, respectively); the combination of the two ligands induced a stronger effect on the differentiation process (anova, P < 0.00001; chi-squared test, P = 4.6 × e−10) (Fig. 1D). Cell-phenotype analysis indicated that after the combined GDNF and ET-3 treatment, cells were strongly induced to differentiate in two main phenotypes (i.e. bipolar and star-like) (chi-squared test, P = 0.034 and P = 9.2 × e−06, respectively). GDNF alone induced cells to differentiate to the star-like phenotype (chi-squared test, P = 2.33 × e−05) (Fig. 1E).

To determine whether or not Apob is modulated by GDNF/ET-3, we treated cells with the two ligands separately and in combination as reported in Fig. 2. We measured Apob mRNA levels through the qRT-PCR and data were normalized using the β-actin gene as endogenous control. GDNF induced an early strong increase in Apob expression after 30 min of stimulation (Fig. 2A), whereas, compared to GDNF, the treatment with ET-3 alone gave a less prominent Apob increment (Fig. 2B). The combination of GDNF and ET-3 led to an increase, although delayed in time, in Apob expression (Fig. 2C).

image

Figure 2.  Effect of GDNF and ET-3 treatment on Apob expression. qRT-PCR of Apob in Neuro2a upon stimulation with GDNF (A), ET-3 (B) and GDNF and ET-3 in combination (C). Data represent the mean (±SE) from three independent experiments.

Download figure to PowerPoint

Apob expression is activated by GDNF/RET signalling pathway

To confirm that the increase in Apob expression was an exclusive effect of Ret activation, we silenced Ret by transient transfection with a Ret-specific short hairpin RNA. Transfected Neuro2a cells were treated with GDNF (50 ng mL−1) at different times (30 min, 1 h, and 3 h), and protein and RNA were extracted. Neuro2a cells silenced for Ret did not show any Apob increase, confirming that the GDNF/Ret pathway activated Apob in a specific manner (Fig. 3A, upper panel). Western blotting confirmed the absence of Ret in the silenced cells (Fig. 3A, lower panel). Transfections with the empty or scramble vector did not abolish Ret-induced Apob increase (data not shown).

image

Figure 3.  Specific activation of Apob transcription by RET/GDNF via MAPK P38 kinase. (A) Ret silencing abolished Apob increased expression in Neuro2a. Upper panel: qRT-PCR for Apob expression in GDNF-treated Neuro2a in GDNF-treated Neuro2a with Ret-silenced (dashed lines). The data represent the mean (error bars) from three independent experiments. Apob mRNA levels are normalized to Αctb. Lower panel: Western blot analysis for Ret protein in serum starved GDNF-treated Neuro2a (first four lanes) and in serum-starved Neuro2a silenced for Ret and treated with GDNF (last 4 lanes). (B) Activation of Apob transcription by RET/GDNF via MAPK P38 kinase. Upper panel: qRT-PCR for Apob expression in GDNF-treated Neuro2a in absence and presence of SB202190 (dashed lines). Lower panel: Western blot analysis of GDNF-treated Neuro2a in absence or presence (last four lanes) of SB202190.

Download figure to PowerPoint

The inhibition of the GDNF/Ret-dependent MAPK P38 kinase with SB202190 completely abolished Apob upregulation (Fig. 3B), whereas Akt and Erk kinase inhibitors did not modify its expression (Figure S2). Therefore, MAPK P38 kinase activation by Ret is responsible for the Apob increase.

Apob silencing reduces HuD expression

These experiments were aimed to understand whether or not Apob activation contributes to ENS development. In this line, we assessed HuD protein changes as an index of neuronal differentiation23 following GDNF treatment of Neuro2a cells transiently silenced for Apob. As shown in Fig. 4A and B, in GDNF-treated Neuro2a cells transfected with the scramble vector (control) the Apob-increased expression was associated with HuD protein increase, an effect prevented by Apob silencing.

image

Figure 4. Apob silencing reduces HuD expression in GDNF-treated Neuro2a cells. (A) Relative Apob expression measured via real-time qRT-PCR in cells transfected with the scramble vector or with the shRNA vector for mouse Apob silencing (shApo) and treated with GDNF at different times. (B) HuD protein levels in Neuro2a cells transfected with scramble vector or with the shRNA vector for mouse Apob silencing (shApo) and treated with GDNF at different times (lower panel). Upper panel: Western blot for vinculin.

Download figure to PowerPoint

Expression analysis of Apob in HSCR mouse models

To establish whether or not GDNF/Ret-induced Apob expression played a role on HSCR pathogenesis, we evaluated Apob expression in a mouse knock-in model for the RetC620R mutation. RetC620R/C620R mice show severe agangliogenesis, kidney agenesis, and die after birth, whereas the heterozygous mice show a normal organ development.19 We tested the expression of Apob in wild-type, heterozygous, and homozygous mouse embryos at day 16 via qRT-PCR (Fig. 5). Compared to wild-type, Apob expression was lower in homozygous mice (Student’s t-test, P = 0.0372), whereas there was a striking increase in expression in heterozygous mice (Student’s t-test, P = 0.0052) as a result of a functioning Ret.19

image

Figure 5. Apob expression in mouse embryos heterozygous and homozygous for Ret C620R (embryonic day 16.5) via qRT-PCR analysis. Data are expressed as means ± SE.

Download figure to PowerPoint

APOB immunolabeling in human colonic tissues

Apolipoprotein B immunoreactivity was identified in cells (with morphological features of immunocytes) of the lamina propria of the adult normal human colon. In contrast, no APOB immunolabeling was detected in enteric neuronal cell bodies or nerve fibers, epithelial cells, or other cell targets of human colon tissue sections (Fig. 6).

image

Figure 6.  Representative examples showing APOB and NSE immunoreactivity in the adult human control colon. (A) Illustrates an intense APOB labeling in cells of lamina propria with features of immunocytes. (B) Shows the lack of APOB immunoreactivity as a control obtained by pre-incubating the primary antibody (1 : 100) with APOB peptide (10−5 m). (C) Shows the lack of APOB immunoreactivity in the myenteric plexus which was clearly identified by NSE labeling (D). Mouse anti-goat and anti-rabbit secondary antibodies conjugated either with fluorescein isothiocyanate (FITC) or tetramethylrhodamine-isothiocyanate (TRITC) were used at 1 : 300 dilution each. Calibration bars: 50 and 100 μm, respectively.

Download figure to PowerPoint

Identification of GDNF/Ret-dependent cis regulatory regions in mouse Apob promoter

To determine the GNDF/Ret-dependent regulatory elements in the mouse Apob genomic region, we first predicted the promoter region in silico using two different softwares: Gene2promoter (http://www.genomatix.de) and PromoterScan. Both programs indicated the region between bp 4983663–4984504 of contig NT_039548 from mouse chromosome 12 as the predicted promoter. We cloned the region upstream of the mouse Apob gene, including the predicted Apob promoter, from genomic DNA and inserted it into the pGL3 vector containing the firefly luciferase gene as reporter. Constructs with progressive deletions of the cloned region were obtained by sequential digestion, and the plasmids were characterized by sequencing (Fig. 7A).

image

Figure 7.  Identification of cis regulatory regions in mouse Apob promoter. (A) Map of the different constructs inserted in the pGL3 basic vector. (B) Results of the luciferase activity for the different constructs, compared to the predicted promoter (900-bp construct). Stars indicate the statistically significant different activities compared to the 900-bp construct. (C) Identification of p53 binding site in mouse ApoB promoter. (D) Luciferase assay of the wild-type and mutant promoter regions (see text for details). (E) EMSA assay showing the binding of p53 (fourth lane, black arrow); gray arrow, complex forming in presence of Neuro2a nuclear extracts.

Download figure to PowerPoint

Each construct was co-transfected into Neuro2a cells with the control vector containing the renilla luciferase gene under the control of the SV40 promoter. One day after transfection, cells were serum starved for 16 h and GNDF stimulation was performed before cell lysis for the luciferase assay. No luciferase activity was found in absence of GDNF (Figure S3). All data were compared to the activity of the 900-bp construct, as this was the minimal promoter region defined by Gene2promoter and PromoterScan. The 1000-bp construct showed a significantly reduced activity compared to the 900-bp construct (P = 0.00020, Student’s t-test), suggesting that the 100 bp extra-region contains repressor elements of Apob expression (Fig. 7B). to identify the cis regulatory elements responsible for Apob GDNF/Ret-dependent transcription regulation, we performed an in silico analysis of this region using MatInspector (http://www.genomatix.de). One site with a high matrix score (thus indicating a real binding site for transcription factors) was identified and represented a binding site for p53 (−986 bp site, matrix score = 0.934) (Fig. 7C). Site-directed mutagenesis of the 1000-bp construct generated a deletion of 2 bp in the p53-binding element (Fig. 7D; upper panel). Activities of wild-type and mutated 1000-bp constructs were compared to the activity of the 900-bp construct via the luciferase assay. The mutated 1000-bp construct showed a comparable activity with the 900-bp construct (P = 0.9457), whereas the original 1000-bp construct activity was reduced (P ≤ 0.0001) (Fig. 7D; lower panel). These data indicate that the p53-binding element acts as a repressor of Apob expression.

We performed an EMSA to verify that p53 actually binds to the region. As shown in Fig. 7E, we first verified that the binding of the Neuro2a nuclear extract to the biotin-labelled p53-putative binding site was specific, as it disappeared in the presence of the unlabelled probe (Fig. 7E, third lane). Next, we were able to detect a supershift when the nuclear extract was pre-incubated with an anti-p53 antibody (Fig. 7E), confirming the binding of p53 to the region.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contribution
  11. References
  12. Supporting Information

RET and EDNRB are critical players during the development of the ENS, as mutations in genes encoding members of either RET or EDNRB pathways lead to a HSCR phenotype. Although several studies have already identified genes that are modulated at the end of each signalling cascade and novel downstream genes activated by GDNF/RET have been identified,24,25 genes modulated by the crosstalk between these two pathways are still matter of investigation.

We have previously identified a vitellogenin (vit-3), corresponding to human APOB, as a differentially expressed gene in C. elegans strains null for the worm homologue of mammalian ECE1.12,26 Several studies have shown the role of APOB in lipid transport and chylomicron production in the gut.13 Nevertheless, APOB mutations have been found in hypobetalipoproteinemia and defective apolipoprotein B-10014,15 and APOB levels have been mainly associated to the risk of chronic heart disease27 and age-related macular degeneration.28 However, a possible correlation between APOB and neuronal function/development arises from studies in Apob homozygous null mice, which displayed neural tube defects and died in utero.16 Thus, the present study was undertaken to investigate whether or not Apob expression has a role in the proliferation/differentiation effects induced by Ret and Ednrb signalling pathways in the Neuro2a cell line, an in vitro model for the study of neuronal development and differentiation.17,18 We have demonstrated that GDNF was a stronger inducer of cell proliferation compared to ET-3. When combined together, ET-3 partially inhibited the proliferative effect of GDNF, in accordance with previous results in enteric neural crest-derived cells committed to neuronal lineage.10 In our cell model, i.e. the Neuro2a cell line, the combination of GDNF and ET-3 induced cells to differentiate into bipolar and star-like phenotypes. Furthermore, GDNF promoted an early increase in Apob expression, which was delayed in combination with ET-3. These data paralleled the effect of GDNF/ET-3 on cell proliferation/differentiation described above, suggesting that the GDNF-mediated early Apob increase may be related to the proliferative effect on neural-crest-derived cells.

We demonstrated that Apob-increased expression was a specific Ret-induced effect due to GDNF stimulation, because Ret silencing abolished Apob upregulation. This increase was dependent on the MAPK P38 kinase pathway, as MAPK P38 inhibition blocks Ret-induced Apob activation. MAPK P38 kinase plays a key role in the balance between cell proliferation and cell death induced by different stimuli.29 To gain further information on the downstream signals activated by Ret via MAPK P38, we evaluated the potential regulatory factors that mediated Apob increase. Although the downstream factors involved in the final regulation of Apob expression are still largely not known, we have proven that p53 is a key regulatory molecule for Apob expression. Luciferase and EMSA assays have shown that p53 acts as a repressor directly binding to the mouse promoter region of Apob. Because MAPK P38 kinase induces p53 phosphorylation,30 it is likely that the GDNF/Ret pathway activates the MAPK P38, which releases p53 from the promoter region thereby inducing Apob transcription. In support of this possibility recent data have indicated that p53 counteracts Ret activity during kidney development.31

Furthermore, we investigated whether or not Apob activation was related to enteric neuronal development by assessing HuD expression in GDNF-treated Neuro2a. Hu proteins are RNA-binding proteins (also referred to as HuD, HuC, HuR, and Hel-N1) that share sequence homology with the embryonic-lethal abnormal vision RNA-binding protein of Drosophila. In particular, HuD is exclusively expressed in neurons, where it exerts a crucial role in neuronal development and survival.32,33 In the present experiments, we provide evidence that HuD protein levels are linked to Apob expression. GDNF treatment of Neuro2a cells transfected with a control vector evoked Apob expression which was associated with HuD increase; in contrast, in Apob silenced Neuro2a cells GDNF did not increase HuD protein. Our data in Neuro2a cells provide support to the concept that Apob can contribute to enteric neuron differentiation and survival via HuD expression. In this line, early data by our group showed that anti-HuD antibodies incubated with primary culture of mature myenteric neurons evoked an apoptotic effect.34

In addition to the in vitro results, homozygous mice knock-in for the HSCR-associated RetC620R had a significant decrease of Apob expression compared to wild-type animals. In contrast, heterozygous mice, which have a normal differentiation and organization of the ENS, displayed a marked increase in Apob expression. The latter effect may be promoted by the wild-type Ret, to compensate for the loss of function of the mutated Ret isoform. In this line, recent data from our group showed that APOB mRNA and protein expression are increased in cells of the immune system and in serum of patients with chronic intestinal pseudo-obstruction compared to healthy controls (Bonora E and De Giorgio R, unpublished observation). The evidence that APOB expression can be detected in these cells is in agreement with data showing RET expression in the lymphoid tissues of the gut, where this receptor exerts an important regulatory function for Peyer’s patch formation.35 Accordingly, in this study, APOB immunostaining was selectively detected in cells, most likely of the immune system, distributed throughout the lamina propria of the normal adult colon.

To our knowledge, this is the first report dealing with Ret-related Apob expression in both in vitro (Neuro2a cells) and in vivo (Ret C620R knock-in mouse) models. More broadly, it may be proposed that RET-dependent APOB expression exerts a major role in regulating different molecular pathways required for enteric neurogenesis and other tissue development. Likewise RET, also APOB may influence the immunoregulatory function in the digestive system, although this exciting aspect clearly deserves further investigation. The data herein presented provide a conceptual framework to evaluate APOB in patients with severe functional gastrointestinal disorders characterized by an impaired motility and transit.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contribution
  11. References
  12. Supporting Information

We thank Dr Kerry J. Rhoden for critical reading and helpful suggestions and Ms Silvia Cerboni for technical help.

Funding

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contribution
  11. References
  12. Supporting Information

This work was supported by FP6 EU grant LSHC-CT-2006-037530 ‘HERMIONE’ and partially supported by FP7 EU grant 223692 ‘CHERISH’ to G. R. This work was also partly supported by grants from the Italian Ministry of University and Research (COFIN Projects to R. De G.), Ricerca Fondamentale Orientata funds from the University of Bologna to R. De G., V. S. and (Centro Interdipartimentale di Ricerca sull’Alimentazione Umana, CIRAU) funds to R. De G. R. De G. is a recipient of grants from Fondazione Del Monte di Bologna e Ravenna, Bologna, Italy. C. E. was a recipient of University of Torino doctoral fellowship. L. M. P. was supported by the Sardininian ‘Master and Back’ project Fellowship.

Author Contribution

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contribution
  11. References
  12. Supporting Information

CE performed the research, analyzed and interpreted the data and wrote the paper; FB, LMP, MR performed the research; AP, IS, AG, MV analyzed and interpreted the data; MS, GR, EB designed the experimental plan and wrote the paper; RDeG and VS revised the whole manuscript and contributed to experimental analysis.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contribution
  11. References
  12. Supporting Information
  • 1
    Gershon MD. Developmental determinants of the independence and complexity of the enteric nervous system. Trends Neurosci 2010; 33: 44656.
  • 2
    Panza E, Knowles CH, Graziano C et al. Genetics of human enteric neuropathies. Prog Neurobiol 2012; 96: 17689.
  • 3
    Amiel J, Lyonnet S. Hirschsprung disease, associated syndromes, and genetics: a review. J Med Genet 2001; 38: 72939.
  • 4
    Arighi E, Borrello MG, Sariola H. RET tyrosine kinase signaling in development and cancer. Cytokine Growth Factor Rev 2005; 16: 44167.
  • 5
    Emison ES, McCallion AS, Kashuk CS et al. A common sex-dependent mutation in a RET enhancer underlies Hirschsprung disease risk. Nature 2005; 434: 85763.
  • 6
    Tahira T, Ishizaka Y, Itoh F, Sugimura T, Nagao M. Characterization of ret proto-oncogene mRNAs encoding two isoforms of the protein product in a human neuroblastoma cell line. Oncogene 1990; 5: 97102.
  • 7
    Arai H, Hori S, Aramori I, Ohkubo H, Nakanishi S. Cloning and expression of a cDNA encoding an endothelin receptor. Nature 1990; 348: 7302.
  • 8
    Carrasquillo MM, McCallion AS, Puffenberger EG, Kashuk CS, Nouri N, Chakravarti A. Genome-wide association study and mouse model identify interaction between RET and EDNRB pathways in Hirschsprung disease. Nat Genet 2002; 32: 23744.
  • 9
    McCallion AS, Stames E, Conlon RA, Chakravarti A. Phenotype variation in two-locus mouse models of Hirschsprung disease: tissue-specific interaction between Ret and Ednrb. Proc Natl Acad Sci U S A 2003; 100: 182631.
  • 10
    Barlow A, de GraaffE, Pachnis V. Enteric nervous system progenitors are coordinately controlled by the G protein-coupled receptor EDNRB and the receptor tyrosine kinase RET. Neuron 2003; 40: 90516.
  • 11
    Kruger GM, Mosher JT, Tsai YH et al. Temporally distinct requirements for endothelin receptor B in the generation and migration of gut neural crest stem cells. Neuron 2003; 40: 91729.
  • 12
    Spanier B, Stürzenbaum SR, Holden-Dye LM, Baumeister R. Caenorhabditis elegans neprilysin NEP-1: an effector of locomotion and pharyngeal pumping. J Mol Biol 2005; 352: 42937.
  • 13
    Rutledge AC, Su Q, Adeli K. Apolipoprotein B100 biogenesis: a complex array of intracellular mechanisms regulating folding, stability, and lipoprotein assembly. Biochem Cell Biol 2010; 88: 25167.
  • 14
    Linton MF, Farese RV Jr, Young SG. Familial hypobetalipoproteinemia. J Lipid Res 1993; 34: 52141.
  • 15
    Soria LF, Ludwig EH, Clarke HRG, Vega GL, Grundy SM, McCarthy BJ. Association between a specific apolipoprotein B mutation and familial defective apolipoprotein B-100. Proc Natl Acad Sci U S A 1989; 86: 58791.
  • 16
    Huang LS, Voyiaziakis E, Markenson DF, Sokol KA, Hayek T, Breslow JL. apo B gene knockout in mice results in embryonic lethality in homozygotes and neural tube defects, male infertility, and reduced HDL cholesterol ester and apo A-I transport rates in heterozygotes. J Clin Invest 1995; 96: 215261.
  • 17
    Lin CH, Hsieh M, Fan SS. The promotion of neurite formation in Neuro2A cells by mouse Mob2 protein. FEBS Lett 2011; 585: 52330.
  • 18
    Sanchez-Ferras O, Coutaud B, Djavanbakht SamaniT, Tremblay I, Souchkova O, Pilon N. Caudal-related homeobox (Cdx) protein-dependent integration of canonical Wnt signaling on paired-box 3 (Pax3) neural crest enhancer. J Biol Chem 2012; 287: 1662335.
  • 19
    Yin L, Puliti A, Bonora E et al. C620R mutation of the murine ret proto-oncogene: loss of function effect in homozygotes and possible gain of function effect in heterozygotes. Int J Cancer 2007; 121: 292300.
  • 20
    Mograbi B, Bocciardi R, Bourget I et al. Glial cell line-derived neurotrophic factor-stimulated phosphatidylinositol 3-kinase and Akt activities exert opposing effects on the ERK pathway: importance for the rescue of neuroectodermic cells. J Biol Chem 2001; 276: 4530719.
  • 21
    Bunone G, Borrello MG, Picetti R et al. Induction of RET proto-oncogene expression in neuroblastoma cells precedes neuronal differentiation and is not mediated by protein synthesis. Exp Cell Res 1995; 217: 929.
  • 22
    De GiorgioR, Volta U, Stanghellini V et al. Neurogenic chronic intestinal pseudo-obstruction: antineuronal antibody-mediated activation of autophagy via Fas. Gastroenterology 2008; 135: 6019.
  • 23
    Liu MT, Kuan YH, Wang J, Hen R, Gershon MD. 5-HT4 receptor-mediated neuroprotection and neurogenesis in the enteric nervous system of adult mice. J Neurosci 2009; 29: 968399.
  • 24
    Heanue TA, Pachnis V. Expression profiling the developing mammalian enteric nervous system identifies marker and candidate Hirschsprung disease genes. Proc Natl Acad Sci U S A 2006; 103: 691924.
  • 25
    Lu BC, Cebrian C, Chi X et al. Etv4 and Etv5 are required downstream of GDNF and Ret for kidney branching morphogenesis. Nat Genet 2009; 41: 1295302.
  • 26
    Bonora E, Evangelisti C, Puliti A et al. Apolipoprotein B is a new target of the RET/GDNF signalling pathway -ESHG, 2011; A-863 (Abstract).
  • 27
    Cromwell WC, Barringer TA. Low-density lipoprotein and apolipoprotein B: clinical use in patients with coronary heart disease. Curr Cardiol Rep 2009; 11: 46875.
  • 28
    Fujihara M, Bartels E, Nielsen LB, Handa JT. A human apoB100 transgenic mouse expresses human apoB100 in the RPE and develops features of early AMD. Exp Eye Res 2009; 88: 111523.
  • 29
    Feng Y, Wen J, Chang CC. p38 Mitogen-activated protein kinase and hematologic malignancies. Arch Pathol Lab Med 2009; 133: 8506.
  • 30
    Brown L, Benchimol S. The involvement of MAPK signaling pathways in determining the cellular response to p53 activation: cell cycle arrest or apoptosis. J Biol Chem 2006; 281: 383240.
  • 31
    Hilliard S, Aboudehen K, Yao X, El-Dahr SS. Tight regulation of p53 activity by Mdm2 is required for ureteric bud growth and branching. Dev Biol 2011; 353: 35466.
  • 32
    Okano HJ, Darnell RB. A hierarchy of Hu RNA binding proteins in developing and adult neurons. J Neurosci 1997; 17: 302437.
  • 33
    Wakamatsu Y, Weston JA. Sequential expression and role of Hu RNA-binding proteins during neurogenesis. Development 1997; 124: 344960.
  • 34
    De GiorgioR, Bovara M, Barbara G et al. Anti-HuD-induced neuronal apoptosis underlying paraneoplastic gut dysmotility. Gastroenterology 2003; 125: 709.
  • 35
    Veiga-Fernandes H, Coles MC, Foster KE et al. Tyrosine kinase receptor RET is a key regulator of Peyer’s patch organogenesis. Nature 2007; 446: 54751.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contribution
  11. References
  12. Supporting Information

Figure S1. Cell morphology classification. Cells were classified according to the presence and shape of neurites as follows: undifferentiated, round-shaped cells (A); with one neurite (monopolar) (B), with two opposite neurites (bipolar) (C); with two opposite and branched neurites (branched-bipolar) (D), with many neurites (star cells) (E); with one prevalent branched neurite (branched) (F).

Figure S2. (A) Western blot analysis of GDNF-treated Neuro2a in absence or presence (last 4 lanes) of Triciribine V, an Akt inhibitor. (B) qRT-PCR for Apob expression in GDNF-treated Neuro2a in presence of Triciribine V. (C) Western blot analysis of GDNF-treated Neuro2a in absence or presence (last 4 lanes) of PD98059, a specific ERK inhibitor. (F) qRT-PCR for Apob expression in GDNF-treated Neuro2a in presence of PD98059.

Figure S3. Luciferase activity assay for the different constructs, compared to the predicted promoter (900 bp construct), in absence of GDNF stimulation.

FilenameFormatSizeDescription
NMO_1998_sm_FigS1-3.docx322KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.