Abstract
Pancreatic islets and acinar tissue develop from duct epithelium and share expression of several transcription factors and other molecular markers also involved with the development of neural tissues. We examined rat pancreatic tissue from fetal life until adulthood for the expression of N-myc downstream regulated gene 4 (Ndrg4), a gene shown to be expressed during neuronal cell differentiation. Isolated pancreatic ducts from neonatal rats were maintained in culture and gave rise to clusters of cells expressing nestin (NES) and PDX-1, which subsequently contained immunoreactive glucagon. Using reverse transcription PCR (RT-PCR), we identified mRNA expression and immunoreactive protein presence for NDRG4 in cultured duct-derived cells, and brain of neonatal rats. By PCR cloning of the ductal cell-derived DNA the molecular form of NDRG4 expressed in pancreatic ducts and ARIP rat pancreatic cells was identified as NDRG4A2, and its presence in intact pancreas of fetal and neonatal rats was demonstrated by immunohistochemistry. Incubation of ARIP cells with glucagon-like polypeptide-1 (GLP-1), increased the expression of NDRG4A2 and PDX-1, while decreasing DNA synthesis and promoting the appearance of glucagon-positive cells. This inhibitory effect of GLP-1 on DNA synthesis and the stimulatory effect on endocrine differentiation were reversed when the translation of NDRG4A2 was prevented using siRNA. These findings indicate that NDRG4A2 is expressed in pancreatic duct cells under GLP-1 control and may be related to a reduction in proliferation and the onset of the pancreas cell differentiation.
Introduction
Both islet endocrine cells and acinar tissue develop from pancreatic epithelium cells during the fetal and neonatal development of the rat, and in the human fetus (Cerf 2006, Murtaugh 2007). The initial development of both lineages depends on the expression of key transcription factors such as PDX-1 and PTF1α within the ductal cells (Murtaugh 2007). PDX-1 is also required in the mature β-cell where it trans-activates the insulin and GLUT2 gene promoters. Other transcription factors including neurogenin3, NEUROG 3, (previously known as NEUROD) BETA2, PAX-4 and -6, and NKX2.2 that are necessary to complete the differentiation of individual endocrine cell lineages, and many of these have a determining role in the differentiation of both neural cells and pancreatic cell types (Habener et al. 2005). Pancreatic ductal cells can be manipulated in vitro giving rise to pseudo-islet structures with multiple endocrine cell types (Ramiya et al. 2000, Soria 2001, Bonner-Weir et al. 2004). Typically, this trans-differentiation process involves the expression of PDX-1. However, there is a paucity of specific markers of pancreatic progenitor cells prior to their expression of endocrine hormones. One such marker was thought to be the intermediate filament protein, the nestin that is also expressed in developing neural tissues (Lendhal et al. 1990). While nestin positive cells were described in isolated adult human islets and identified as a sub-fraction expressing the ATP-binding cassette transporter (ABCG2; Lechner et al. 2002), the corresponding cells in vivo are usually mesenchymal in nature and have been linked to the development of new endothelial cells within the supporting vasculature (Treutelaar et al. 2003). This does not preclude nestin expression also being associated within the lineage development of some pancreatic endocrine or acinar precursor cells (Delacour et al. 2004, Joanette et al. 2004).
We therefore hypothesized that there might be other genes co-expressed in developing pancreas and brain that might be involved in pancreatic cell differentiation from ductal cells. One such candidate is the N-myc downstream regulated gene 4 (Ndrg4)/Bdm1 family. These genes are expressed throughout regions of the brain in the neonatal rat and are required for neurite outgrowth (Yamauchi et al. 1999, Ohki et al. 2002). A mutant Ndrg1 gene is the cause of hereditary motor and sensory neuropathy-Lom, marked by progressive axonal loss on the peripheral nervous system. All Ndrg proteins contain an α/β hydrolase structure common to hydrolytic enzymes (Shaw et al. 2002). Here, we describe the identification of NDRG4A2 in rat pancreatic duct cells, and show a possible association to pathways of endocrine cell differentiation.
Materials and Methods
Pregnant Wistar rats were purchased from Charles River Ltd, St Constant, QC, Canada. Animals were maintained under standard conditions and had access to rodent chow and water ad libitum. All procedures in this study had the prior approval of the animal care committee of the University of Western Ontario and were performed in accordance with the guidelines of the Canadian Council for Animal Care. Collagenase, fetal bovine serum, 100 bp and 123 bp DNA ladders, Taq DNA polymerase, and Superscript II RNAse H reverse transcriptase, siRNA, DMEM, Ham's F-12 and F12K medium were each purchased from GIBCO BRL. RQ1 DNase was purchased from Promega and QIA Quick Gel Extraction Kit from QIAGEN Inc. Trizol reagent, Taq DNA polymerase, oligo dT primers and reverse transcriptase were obtained from Invitrogen Inc. Mouse anti-nestin monoclonal antibody was purchased from Chemicon International Inc., Temecula, CA, USA while rabbit anti-human NDRG-4 was kindly provided by Dr K Kokame, National Cardiovascular Center Research Institute, Osaka, Japan. Monoclonal antibody against human NDRG4 was obtained from the Abnova Corporation, Taipei, Taiwan together with recombinant NDRG4 peptide for antibody blocking studies. Rabbit-anti-porcine glucagon was a gift from Dr Ungar, Dallas, TX, USA and guinea pig anti-insulin was kindly provided by Dr T J McDonald, University of Western Ontario, London, ON, Canada. Mouse anti-human cytokeratin 20 and monoclonal antibody against endothelial cell CD31/PECAM-1 were obtained from Dako Corporation, Santa Barbara, CA, USA, and rabbit anti-human α amylase from Sigma Chemical Co. Rabbit anti-rat PDX-1 was provided by Dr C Wright, Vanderbilt University. Glucagon-like polypeptide 1 (GLP-1) and all other reagents were purchased from Sigma.
Isolation and culture of pancreatic duct cells
Neonatal rats were killed by decapitation on postnatal day 4 and the pancreas, brain (cerebral cortex), and duodenum removed into Hank's buffered salts solution (HBSS) or frozen under liquid nitrogen for RNA isolation. Pancreatic ducts were isolated by a modification of the methods described originally by Remiya et al. (2000). Briefly, pancreata were dissected and pooled from 8–10 animals and digested with collagenase (2.8 mg/ml) in HBSS (5 ml). Tissue digestion was performed for ∼10 min at 38 °C in a shaking water bath (200 cycles/min). The dissection was terminated by the addition of 10 ml ice-cold HBSS, the contents dispersed by pipette, and the tissue collected by centrifugation at 1000 g for 10 min. Two further washes with ice-cold HBSS were performed. Elongated ductal structures were hand-picked under a dissecting microscope and cultured in Ham's F-12/DMEM (50/50) with 2% fetal bovine serum supplemented with 10 000 U/ml penicillin and 50 μg/ml streptomycin in tissue-culture grade plastic dishes (Falcon, Lincoln Park, NJ, USA) at 37 °C in a humidified atmosphere of 95% O2/5% CO2. The yield of ducts was ∼150 per 10 animals. Ductal structures were allowed to attach to the culture dishes and medium was changed every 4 days. After 2 weeks culture, monolayers of cells had formed from the ducts and spherical clusters of cells budded from these, became detached and floated within the culture medium. These were collected and sub-cultured to be further analyzed for RNA or protein presence. Immunocytochemical staining with antiserum against cytokeratin 20 demonstrated that over 90% of the original monolayers were immunopositive, while no staining was seen for α amylase or for CD31 as a marker for endothelial cells deriving from capillaries. Less than 1% of cells in the original monolayers were immunopositive for insulin or glucagon. The positive staining for cytokeratin, coupled with the absence of endocrine hormone-positive cells, acinar or endothelial cells suggests that the cultures consisted predominantly of ductally-derived epithelial cells.
RNA extraction and reverse transcription PCR (RT-PCR)
To detect the expression of specific mRNAs within duct-derived or ARIP cell cultures, RT-PCR was used as previously described (Wang et al. 2000). Briefly, total RNA was extracted from duct-derived epithelial cells, newly isolated ducts, neonatal rat brain or duodenum using Trizol Reagent (Invitrogen). Purified RNA (100 μg) was extracted using the RNeasy Mini Kit (Qiagen) to remove any contamination of genomic DNA, and the integrity was verified by the separation using gel electrophoresis and visualization by ethidium bromide staining. The amount of RNA was estimated by absorbance at 260 nm, and samples were stored at −80 °C. Three μg total RNA was reverse transcribed with 300 U of SuperScript II Rnase H− reverse transcriptase (Invitrogen) using 0.5 μg oligo (dT)12–18 primers (Invitrogen) according to manufacturers instructions. The primers utilized are shown in Table 1, and included those designed to amplify a region of the open reading frame of rat Ndrg4 (sequence 181–480). After a denaturation step of 3 min at 95 °C, 0.5 μl Taq DNA polymerase (Invitrogen) was added at 80 °C to perform a hot-start. Following seven cycles of 1 min at 94 °C for denaturation, 1 min at 55 °C for annealing and 1 min at 72 °C for extension, 2 μl β-actin primers (50 μM) were added to complete 35 more cycles and extension for the last 10 min at 72 °C. To establish a linear range of amplification, several different cycle numbers of PCR (20, 25, 30, 35, 40, 42, 45) were run. Negative controls without reverse transcriptase (-RT), and without cDNA (-PCR) were run, and no genomic DNA contamination was found. The β-actin primers covered exon/intron boundaries, and no signal at 970 bp of genomic DNA contamination was found (data not shown) compared with the expected product size of 660 bp. The amplified PCR products were visualized on 2% agarose gels stained with ethidium bromide and analyzed using a gel imaging system (AlphaEaseFC-FluorChem 8800 software, Alpha Innotech Corp., San Leandro, CA, USA). For the identification and cloning of Ndrg4, individual bands were removed from the gels, purified by QIA Quick Gel Extraction Kits, and the contained cDNA fragments extracted and re-amplified by PCR using the same primers. The DNA was sequenced within the University of Western Ontario core facility and the sequences identified through the NIH Gene Bank.
Primer sets used for the identification of mRNAs encoding N-myc downstream regulated gene 4 (Ndrg4), Pdx-1, nestin, N-myc, insulin, glucagon, and actin in pancreatic ductal cell and ARIP cell cultures
| Ndrg4 antisense: | TCA, ATG, TTC, ATC, AAC, ACC, AGC |
| Sense: | TCT, CAA, TCA, CAA, GCT, GTG, CTT |
| Pdx-1 antisense: | CGC, TGA, GAG, TCC, GTG, AGC, TGC |
| Sense: | TGC, AAA, CTG, GGG, TGG, CGG, CGG |
| Nestin antisense | TGG, TGA, GGG, TTG, AGG, TTT, GT |
| Sense: | AGG, CTT, CTC, TTG, GCT, TTC, TGG |
| N-myc antisense: | CGA, TGT, GGT, CAC, GGT, AGA, GA |
| Sense: | AAG, TCC, GAG, CGT, GTT, CGA, TCT |
| Insulin antisense: | TGC, CCA, GGC, TTT, TGT, CAA, ACA, GCA, CCT, T |
| Sense: | CTC, CAG, TGC, CAA. GGT, CTG, AA |
| Glucagon antisense: | GTG, GCT, GGA, TTG, TTT, GTA, ATG, CTG |
| Sense: | CGG, TTC, CTC, TTG, GTG, TTC, ATC, AAC |
| Actin antisense: | GCC, CCT, CTG, AAC, CCT, AAG |
| Sense: | CAT, CAC, AAT, GCC, AGT, GGT, A |
Cloning of Ndrg4A2
PCR cloning was used to identify the entire cDNA sequence of the gene Ndrg4A2 gene expressed within pancreatic duct-derived epithelial cells and in the rat ARIP pancreatic duct cell line. RNA was extracted from cells and digested by DNAase as described above. RT-PCR was performed as described above and the individual bands were excised and sequenced. Identified sequences were compared with those published for Bdm1 and Ndrg4A2. To identify the 3′ end and 5′ end of Ndrg4A2 mRNA the RACE method was used as described by Frohman et al. (1988). Purified RNA (200 ng) was used as the template in a reverse transcription reaction, using the anchor primer 5′ ACG, TAA, TAC, GAC, TCA, CTA, TAG, GGC, GAA, TTG, GGT, CGA, C-d (T)18A-3′ to identify the 3′ end of the Ndrg4B cDNA. Reactions without reverse transcriptase were used as the negative control. The Ndrg4A2 gene-specific primer (CGG, GTC, CTT, CTG, TAT, GTG, GTT, TCC) and the anchor primer were used in PCR and the products separated on 2% agarose gel and subjected to sequence analysis. To identify the 5′ end mRNA of Ndrg4A2, the purified RNA and Ndrg4A2 gene-specific primer (TCA, ATG, TTC, ATC, AAC, ACC, AGC) was used in the reverse transcription. cDNA (5 μl), 5× TdT buffer (2 μl), 15 mM CoCl2, (1 μl) 1 mM dATP (1 μl), and terminal transferease (1 μl, 25 U) were mixed and incubated at 37 °C for 30 min. The reaction was purified on a QIA column and the eluate (30 μl) mixed with 10 Tag buffer (5 μl), 2.5 mM dNTP (3 μl), 20 μM (dT)17 adaptor (GAC, TCG, AGT, CGA, CAT, T17; 1 μl), and water (11 μl), and incubated at 95 °C for 5 min; 72 °C for 30 s while adding 2 U of Tag polymerase; 40 °C for 4 min; and at 72 °C for 40 min. The product was subjected to PCR using the Ndrg4A2 specific primer (CAT, GGC, ACA, CCA, CGA, AGT, GTT, TG) and the adaptor primer (GAC, TCG, AGT, CGA, CAT, CG) and the cDNA fragment separated and sequenced.
Immunohistochemistry
Cell clusters derived from ductal cultures were collected and washed three times with PBS. The cells were spread onto glass cover slips by centrifugation for 5 min at 500 g and air-dried at 23 °C before fixation with 70% (v/v) ethanol. Immunohistochemistry was performed on cells and on paraffin sections (5 μm) of rat pancreas (postnatal days 8, 21, and 130) or cerebral cortex (day 8) using antibodies against nestin (1:200 dilution in PBS), insulin (1:15 dilution), glucagon (1:100 dilution), cytokeratin 20 (1:50 dilution), α amylase (1:2000 dilution), PDX-1 (1:2500 dilution), CD31/PECAM-1 (1:50 dilution) or human NDRG4 (1:100 dilution) by the avidin-biotin peroxidase method as previously described (Joanette et al. 2004). Biotinylated horse anti-mouse, goat anti-rabbit, and goat anti-guinea pig (1:100 dilution, Vector Laboratories, Burlingame, CA, USA) were used as secondary antibodies. After washing in 50 mM Tris–HCl (pH 7.5), peptide immunoreactivity was localized by incubation in fresh 1.89 mM diaminobenzidine with 0.03% (v/v) hydrogen peroxide for 2 min and the reaction was quenched in excess 50 mM Tris–HCl (pH 7.5). Cells or tissue sections were counter-stained with Carazzi's hematoxylin, dehydrated and mounted under glass cover slips. Reactions in the absence of either primary or secondary antibody or the use of non-immune serum, served as negative controls. As an additional control of specificity the monoclonal antibody against NDRG4 was pre-absorbed overnight at 4 °C with 100 μM NDRG4 peptide, any precipitate removed by centrifugation, and the supernatant used in the place of the primary antibody. Immunohistochemistry for NDRG4 was also performed on sections of liver and duodenum from 8 day-old animals for comparison with pancreas.
Western blot analysis
Proteins were extracted from ARIP cell cultures using 1 ml lysis buffer (20 mM Tris–HCl (pH 7.4), 1% Triton X-100, 10% glycerol, 150 mM NaCl, 2 mM EDTA, 25 mM β-glycerophosphate, 20 mM sodium fluoride, 1 mM sodium orthovanadate, 2 mM sodium pyrophosphate, 20 μM leupeptin, 1 mM benzamidine, 1 mM 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride, 1 μM microcystin) for ∼2×105 cells at 4 °C for 30 min. Detergent-insoluble material was precipitated by centrifugation at 12 000 g for 10 min at 4 °C and lipid layers were removed. Whole protein concentrations from the supernatant were determined by the Micro BCA protein assay (Pierce, Rockford, IL, USA) using BSA as standard protein, as per manufacturer directions. Protein samples were diluted in the SDS sample buffer, and the mixture was boiled for 5 min. Proteins (20 μg) were separated on an 8% SDS/PAGE and then transferred to nitrocellulose membranes. After blocking the membranes for 60 min at room temperature in Tris buffered saline containing 0.05% Tween 20 (TTBS) and 5% non-fat dry milk or 5% BSA, they were probed overnight at 4 °C with human Ndrg4 antibody (1:100 dilution). Membranes were washed in TTBS and the proteins detected by enhanced chemiluminescence (Pierce) with HRP-labeled anti-mouse secondary antibody.
Regulation of Ndrg4A2 in ARIP cells
Rat ARIP pancreatic duct cells were maintained in F12 medium containing 7 mM glucose and 10% FCS. Cells were plated in culture dishes (1.7×104 per well) and maintained until 70% confluent. They were subsequently cultured in serum-free medium containing 12 mM glucose and increasing concentrations of GLP-1 for 48 h. [3H]-thymidine (2 μCi/ml) was added to some cultures and the isotope incorporated into DNA measured as previously described (Wang et al. 2000). RNA was extracted from ARIP cells and analyzed by RT-PCR as described above. To examine the effect of the Ndrg4A2 gene on the regulation of DNA synthesis and the expression of Pdx-1 in ARIP cells, RNA interference was used as described by Gitlin et al. (2002). Briefly, the specific sequence (CCA, AGT, TTG, CAC, TCA, TCT, T) targeted on the open reading frame of Ndrg4A2 gene were chosen and siRNA was synthesized by Invitrogen, together with a non-specific control siRNA of equivalent size designed to have no significant sequence similarity to known mouse, rat or human transcript sequences. Ndrg4A2 siRNA or control RNA (40 pmoles/ml) was pre-incubated with serum-free medium for 20 min at 23 °C before addition to cell cultures and incubation for 4 h at 37 °C. Transfection efficiency was estimated to be in excess of 80% using a control transfection with Ambion Silencer Cycle 3 GFP siRNA (Applied Biosystems, Foster City, CA, USA) and visualization of GFP expression in cells after 24 h by fluorescence microscopy. The culture medium was replaced by fresh medium with the addition of GLP-1 (0–20 nM) and [3H] thymidine (2 μCi/ml) for 48 h at 37 °C. Thymidine incorporation was quantified and RNA extracted and analyzed by RT-PCR as described above. To determine the involvement of Ndrg4A2 in ARIP cell differentiation, cells were incubated in F12 medium containing 7 mM glucose and 1% FCS for 14 days in the presence of GLP-1 (20 nM), with or without Ndrg4A2 siRNA. Cells were then fixed and subjected to immunocytochemistry for glucagon or insulin as described above.
Statistical analysis
Experiments with pancreatic ductal cells, or with the ARIP cell line were performed with three replicate culture plates per variable, and each experiment repeated between three and five times. Data were analyzed by multivariate ANOVA. Immunohistochemistry was performed on tissues derived between three and five separate animals and representative views are shown. To calculate the percent of primary ductal cells immunopositive for both nestin and PDX-1 following duel immunhistochemistry, all cells demonstrating cytoplasmic staining for nestin within a single culture dish were quantified using a Carl Zeiss transmitted light microscope at a magnification of ×40. The percent of immunopositive cells that also demonstrated nuclear immunolocalization of PDX-1 was then calculated. The number of ARIP cells that demonstrated cytoplasmic immunoreactivity for glucagon following incubation in the presence of GLP-1 was quantified relative to the total number of cells per culture flask. Microscopic analyses for cell number were performed with Northern Eclipse (version 6.0) morphometric analysis software (Empix Imaging, Mississauga, ON, Canada).
Results
Isolation and culture of pancreatic ductal cells
Isolated rat pancreatic ducts placed in culture initially formed elongated structures that resembled ducts but had no detectable lumen (Fig. 1A). After culture for two weeks, monolayers of cells had grown out from these and the original ductal structures were no longer discernible. Spherical cell clusters were seen to bud out from the monolayers after three weeks of culture and often detached into the medium. Such clusters, but not the original monolayers were immunopositive for the presence of nestin, a possible marker of neural and pancreatic cell precursor development (Fig. 1C). Approximately 30% of nestin-positive cells also demonstrated nuclear staining for PDX-1 (Fig. 1C). Clusters were collected, enzymatically dispersed, and re-plated and after 3 weeks groups of cells demonstrated immunoreactivity for glucagon (Fig. 1B), but not insulin, α amylase or CD31. Messenger RNAs for nestin and for Pdx-1 were expressed within these cultures when analyzed by RT-PCR (Fig. 1E and F).
Histological appearance of pancreatic ducts 24 h after isolation and plating onto tissue culture dishes (A), and the appearance of spherical cell clusters derived from the subsequent monolayers of cells and sub-cultured (B–D). Cell clusters demonstrated immunoreactivity for glucagon (B), nestin (C), and NDRG4 (D) (arrows). Many cells immunoreactive for nestin showed nuclear co-staining for PDX-1 (dark blue) (C). Magnification bar A −50 μm, B, C and D −20 μm. The presence of mRNA for nestin (E) or PDX-1 (F) within duct-derived cell clusters was determined by RT-PCR analysis (lanes B and C). The positive control for nestin consisted of RNA from neonatal rat brain, and for PDX-1, newly isolated pancreatic ducts (lanes D and E). In both panels, reactions without added template (lane A) or without reverse transcriptase (lanes C and E) were used as controls. The molecular markers (M, bp) are shown on the left.
Citation: Journal of Endocrinology 201, 1; 10.1677/JOE-08-0296
Identification and cloning of Ndrg4A2 gene in pancreatic duct cells and brain
We sought to identify NDRG4 expression within pancreatic duct-derived cell clusters and identified a positive signal using RT-PCR for RNA isolated from both ductal cells and neonatal rat brain. The sequence of the Bdm1/Ndrg4 mRNA expressed in brain or ductal cells was determined by PCR cloning using different primer pairs (Fig. 2A), while the 3′ and 5′ ends of Ndrg4A2 mRNA expressed in the duct cells were identified by RACE. The identified sequences of mRNA expressed in duct-derived cells were aligned with the sequences of Bdm1 (GI 4105411) and Ndrg4A2 (GI 37788064) published in Gene Bank using the pair-wise blast program. The cDNA sequence detected in duct-derived cell clusters was identical to that of Bdm1, except for the deletion of 39 nucleotides from positions 926–965 in the open reading frame (Fig. 2B), and was identified as Ndrg4A2, an isoform with a transcript starting from exon 5 and lacking exon 18 of the gene (Maeda et al. 2004).
(A) Primer sequences used for the PCR cloning of Ndrg4A2 expressed in pancreatic ductal cells. According to the published sequence of Ndrg4A2 mRNA, various primer pairs were designed as indicated by letters (A–F) and used in RT-PCR analysis for ductal cells or neonatal rat brain tissue as a positive expression control. The triangles indicate the location of introns and the bold line indicates the location of the open reading frame of the gene. (B) Comparison of cDNA sequences for Ndrg4A2 expressed in the neonatal rat brain and that identified in cultured pancreatic ductal cells. Sequences identified by cDNA cloning and RACE were aligned with the published sequences of the NdrgA4 (Bdm1) gene. A sequence with 39 bp and 13 amino acids is missing in the open reading frame of Ndrg4A2 expressed in ductal cells.
Citation: Journal of Endocrinology 201, 1; 10.1677/JOE-08-0296
Using immunocytochemistry with two separate antibodies, we identified NRDG4 within duct-derived cell clusters (Fig. 1D), and determined that 43±4% of cells also demonstrated immunoreactivity for PDX-1. Immunohistochemistry was also utilized to show the presence of NRDG4 within the intact pancreas of fetal (gestational day 20; Fig. 3A) and neonatal rats (postnatal day 8; Fig. 3B), predominantly associated with the pancreatic duct epithelium and with acinar cells, but not within islet cells. Pancreas isolated from animals at 21 days postnatal age showed only slight staining for NDRG4 within the ducts (Fig. 3C), while by 130 days age no immunoreactivity was detected (Fig. 3D), suggesting that the presence of NDRG4A2 was absent after weaning. Staining in the pancreatic ducts for Ndrg4 was co-localized with cytokeratin (Fig. 4A and B). Staining for NDRG4 was completely absent in control conditions in which the primary antibody was omitted or pre-absorbed with excess antigen (Fig. 3F). No immunohistochemical localization of NDRG4 was seen in two other neonatal tissues derived from gut endoderm, liver or duodenum (not shown), but positive staining was seen within a sub-population of neurons in the cerebral cortex of the neonatal rat (Fig. 3E).
Localization by immunohistochemistry of NDRG4 within the pancreas of the fetal (gestational day 20) (A), or neonatal rat (8 days postnatal) (B), at day 21 postnatal (C), or as adults at 130 days postnatal (D). Positive staining was seen within cells of the pancreatic duct epithelium (arrows) and within acinar cells (a), but not within islets (I), up until postnatal day 8. The presence of NDRG4 was much reduced in pancreas at 21 days and was absent at 130 days. Staining was also shown within the cerebral cortex from the same animals as in B at postnatal day 8 (E), being localized to a sub-population of neurons (arrows). When the antibody to Ndrg4 was pre-absorbed with excess antigen the ductal and acinar staining seen at day 8 was abolished (F). Magnification bar 100 μm.
Citation: Journal of Endocrinology 201, 1; 10.1677/JOE-08-0296
Localization by immunohistochemistry of cytokeratin (A) or NDRG4 (B) within adjacent sections of the pancreas of the fetal rat (gestational day 19). Positive staining was seen for both ligands within cells of the pancreatic duct epithelium (amplified inset). Magnification bar 50 μm.
Citation: Journal of Endocrinology 201, 1; 10.1677/JOE-08-0296
Expression and regulation of Ndrg4A2 in ARIP cells
The expression and sequence of Ndrg4 expressed in the ARIP rat pancreatic duct cell line was determined by RT-PCR cloning and found to be identical to Ndrg4A2 identified in primary ductal cultures. The effect of RNA interference on blocking translation of Ndrg4A2 was examined by RT-PCR analysis. NDRG4A2 expression was increased when the ARIP cells were incubated with 20 nM GLP-1 (Fig. 5). However, after addition of siRNA coding for a specific sequence of Ndrg4A2 the expression of the longer cDNA fragment of Ndrg4A2 (181–681) was abolished and the abundance of the short cDNA fragment of the gene (181–481) was substantially decreased. The expression of NDRG4A2 was not affected when a non- specific siRNA was used, and actin expression was unaltered. These observations indicate that the effect of the siRNA was specific.
The effect of siRNA interference for Ndrg4A2 (sequence location shown) on the expression of mRNAs encoding NDRG4A2 and actin. ARIP pancreatic duct cells were cultured in serum-free medium with (lanes 2–4) or without (lane 1) 20 nM GLP-1 after prior transfection with a specific siRNA for Ndrg4A2 (lane 3) or non-specific RNA (lane 4). The mRNAs were quantified using RT-PCR using separate primers for short and long sequences for Ndrg4A2. The analysis of actin mRNA in the same preparation was used as the loading control and to exclude possible contamination by DNA. A reaction without adding primers was used as the negative control (C).
Citation: Journal of Endocrinology 201, 1; 10.1677/JOE-08-0296
We then examined if changes in NDRG4A2 expression following exposure of ARIP cells to GLP-1 were associated with the reported induction of differentiation into endocrine or acinar phenotypes (Bulotta et al. 2002). Exposure to GLP-1 was associated with a significant reduction in DNA synthesis, as determined by the incorporation of [3H] thymidine (Fig. 6A). The inhibitory effect of GLP-1 on [3H] thymidine incorporation by ARIP cells was rescued when siRNA specific for Ndrg4A2 was applied to cells (Fig. 6A), while the addition of a control non-specific RNA did not significantly affect thymidine incorporation. Cells transfected with control siRNA showed immunoreactivity for NRDG4, as did non-transfected cells (Fig. 6B-a,c), which was absent in cells exposed to siRNA specific for Ndrg4A2 (Fig. 6B-b). Furthermore, western blot analysis of ARIP cell proteins revealed a 35 kDa species in non-transfected cells, or those transfected with control siRNA, which was substantially less abundant after transfection of cells with siRNA for Ndrg4A2 (Fig. 6C).
(A) The effect of GLP-1, with or without siRNA inhibition of NDRG4A2 synthesis, on [3H] thymidine incorporation (c.p.m./well) by ARIP pancreatic duct cells, (B) the presence of immunoreactive NDRG4 (a–c and f) or glucagon (d and e), and (C) the presence of NDRG4 protein assessed by western blot analysis following transfection with control siRNA or siRNA for Ndrg4A2. In (A) cells were pre-incubated without siRNA (1–4), with a non-specific RNA control (5 and 7) or with the addition of siRNA specific for Ndrg4A2 (6). Cells were then incubated in serum free medium without (1 and 5), or with GLP-1 at 5 nM (2), 10 nM (3) or 20 nM (4, 6, and 7) concentrations. The addition of 20 nM GLP-1 significantly inhibited thymidine incorporation (1 versus 4, 5 versus 7, *P<0.05; n=3). This inhibitory effect was rescued by transfection with siRNA for Ndrg4A2 (6 versus 7, #P<0.05; n=5). (B) ARIP cells were subjected to immuncytochemistry to visualize NDRG4 following transfection with control siRNA (a), siRNA for Ndrg4A2 (b) or no siRNA (c). Immunoreactive NDRG4 was visible in control cultures (arrows) but was absent in cells exposed to NDRG4A2 siRNA, confirming an inhibition of protein translation. To determine the effects of NDRG4A2 on differentiation, ARIP cells were incubated for 14 days in the presence of GLP-1 (20 nM) with control siRNA (d), siRNA for Ndrg4A2 (e) or without transfection (f). Cells were then subjected to immunocytochemistry for glucagon (arrows). Cells transfected with siRNA encoding Ndrg4A2 had a lower frequency of glucagon immunoreactivity. Magnification bar 20 μm. (C) ARIP cells were cultured without transfection, following transfection with control siRNA, or following transfection with siRNA for Ndrg4A2, protein extracted and a western blot performed for NDRG4 protein. A single species of ∼35 kDa was visualized in control cultures but was substantially reduced following transfection with siRNA for Ndrg4A2.
Citation: Journal of Endocrinology 201, 1; 10.1677/JOE-08-0296
ARIP cells were incubated in the presence of 20 nM GLP-1 for 14 day with or without siRNA for Ndrg4A2. In the presence of GLP-1 alone or with control siRNA, foci of glucagon-immunoreactive cells were seen (Fig. 6B–d, f; 16±4% of cells), which were absent in control cultures. The presence of glucagon-immunoreactive cells was significantly reduced in the presence of siRNA for Ndrg4A2 (Fig. 6B-e; 5±2% of cells, P<0.01 versus GLP-1 alone), suggesting that inhibition of translation of NDRG4A2 prevented GLP-1-induced differentiation of glucagon+ve endocrine cells. No immunoreactive insulin was detected in ARIP cultures after 14 days exposure to GLP-1.
The expression of NDRG4A2 was inversely related to cell DNA synthesis, being increased in response to GLP-1 (Fig. 7), as was the expression of N-myc, while the expression of the housekeeping gene actin was unchanged. Messenger RNA for PDX-1, an indicator of ductal cell differentiation into endocrine or acinar cell lineages, was also increased in response to GLP-1 (Fig. 7).
Expression of mRNAs encoding Pdx-1, N-myc, Ndrg4a2, and actin within ARIP pancreatic ductal cells by RT-PCR following treatment with increasing concentrations of GLP-1 in serum-free medium (0% FC). The reaction without added primers was used as the negative control (C). The analysis of actin in the same preparations of mRNA was used as the loading control and to exclude the possibility of contamination with DNA.
Citation: Journal of Endocrinology 201, 1; 10.1677/JOE-08-0296
Cell clusters derived from primary cultures of neonatal rat pancreatic ducts were also cultured in the presence of GLP-1 (20 nM) for up to 3 weeks. The number of cells found to be immunoreactive for glucagon was slightly increased (control 13±2%, GLP-1 18±2%, P<0.05, n=4 preparations).
Discussion
We have identified the expression of a member of the Ndrg gene family, Ndrg4A2, within pancreatic ducts and shown that this could be involved in the regulation of pancreatic cell differentiation from ductal precursors in response to GLP-1. Previous studies found that Bdm1 was homologous to human Ndrg 4 (Yamauchi et al. 1999, Nakada et al. 2002), one of four known members of the Ndrg family. Ndrg4 has three isoforms, designated Ndrg4B, Ndrg4BVar, and Ndrg4H. These isoforms have an identical 3′- untranslated region (UTR) sequence, while the nucleotide sequences of the 5′- ends of Ndrg4B and Ndrg4BVar are identical but differ from Ndrg4H. The sequences of Ndrg4B and Ndrg4BVar are identical except for a deletion of 39 bp in the open reading frame of Ndrg4B from bp 927 to bp 965 due to alternative RNA splicing. (Zhou et al. 2001, Qu et al. 2002). After cloning the NDRG expressed in pancreatic ductal cells, we identified the deletion of 39 bp characteristic of Ndrg4B, suggesting that this was the rat homolog. A recent study has detailed the genomic structure of rat Ndrg4 (Maeda et al. 2004). Six transcripts of the gene have been found and named as NDRG4A1, A2, B1, B2, C1, and C2. The transcripts derived from the 5′-ends of exons 1, 2, and 5 are referred to as NDRGC, NDRG4B, and NDRG4A respectively. Transcripts that contain exon 18 of Ndrg4 are denoted NDRG4A1, NDRG4B1, and NDRG4C1, and those lacking exon 18 are Ndrg4A2, B2, and C2. NDRG4A2, as found here in rat pancreas, is homologous to NDRG4B (Maeda et al. 2004).
We identified NDRG4A2 in duct cells of neonatal rat pancreas, in cells derived from ductal cultures with the potential to develop along an endocrine lineage, and in the ARIP rat pancreatic cell line which can give rise to both acinar and endocrine cells (Hui et al. 2001, Liu et al. 2004). Previous studies had found that the expression of NDRG4 was restricted to neuronal tissue, the expression of NDRG4A1 and A2 being mainly found in rat brain shortly after birth at a time of neural plasticity (Nakada et al. 2002). NDRG4 was also found to be expressed in heart, but not in other rat tissues examined such as spleen, lung, liver, skeletal muscle, kidney, and testis. It was detected in undifferentiated rat PC12 pheochromocytoma cells but was not detected in eight other non-neural cell lines examined (Yamauchi et al. 1999). In human, NDRG4B is expressed in brain, NDRG4H in brain and heart, while NDRG4BVar is only marginally expressed in human tissue (Zhou et al. 2001, Qu et al. 2002). The expression of NDRG4A2 in rat pancreas indicates yet another gene shared between pancreas and brain tissues that might be involved with tissue maturation, as has been found for neurogenin3, NEUROD, and nestin (Gasa et al. 2004). However, the RNA processing of NDRG4 in these two tissues could be different allowing for tissue specific actions.
The expression of NDRG4 in rat brain is linked to neuronal developmental processes, and in P19 murine embryonic carcinoma cells is increased after the induction of neuronal differentiation (Bulotta et al. 2002). The transfection of NDRG4 into PC 12 neuronal cells regulated the outgrowth of neurites and the expression of the AP-1 DNA binding protein (Yamauchi et al. 1999, Ohki et al. 2002). By analogy, the expression of NDRG4 in pancreas could also be related to growth and differentiation. Pancreatic duct epithelium is able to function as a precursor pool for cells that can contribute to both endocrine and exocrine lineages in vitro, and through budding of cells from the ducts during embryonic development (Githens 1988, Bonner-Weir & Weir 2005). Mouse pancreatic tissue has been shown to contain multi-potential stem cells at low frequency that are capable of giving rise to all pancreatic endocrine cell lineages, as well as neural cell types (Seaberg et al. 2004). It is also likely that β-cells can de-differentiate to a ductal phenotype, and subsequently re-differentiate back to endocrine cell types (Wang & Yashpal 2004). In the neonatal rat pancreas, NDRG4A2 was present within pancreatic duct cells and in the developing acinar tissue, but was absent from endocrine cells within the islets of Langerhans. The localization to acinar tissue was substantially reduced by postnatal day 21, and no presence of NDRG4A2 was detected in adult pancreas. The temporal distribution is consistent with a role of NDRG4A2 in the differentiation of acinar tissue, which develops rapidly in the neonatal rodent, and suggests that the gene is no longer expressed once differentiated endocrine cells are present within the islets.
We found that ductal cultures from the neonatal rat pancreas generated cell clusters that were proliferative and expressed the intermediate filament protein, nestin. Nestin-positive cells seen in the neonatal pancreas are likely to predominantly represent cells contributing to the vascular endothelium (Joanette et al. 2004), although there are some reports also associating nestin with cells within the endocrine lineage (Maria-Engler et al. 2004). The duct-derived cell clusters could be sub-cultured for up to 6 months and expressed the early transcription factor marker of endocrine and acinar cell commitment, PDX-1. Similar results were found for islet progenitor cells that budded from cultured ductal monolayers derived from adult NOD mice (Ramiya et al. 2000). We found no expression or presence of insulin within the ductal cell cultures, although they did demonstrate immunoreactive glucagon presence after 3 weeks of culture. During embryonic development of the mouse, cells containing glucagon and tyrosine hydroxylase are the first to appear in the pancreatic duct at day e10 (Alpert et al. 1988).
The ARIP rat pancreatic cell line is a useful tool since these cells can differentiate into endocrine and acinar lineages. Following exposure to GLP-1 they were reported to give rise to insulin-expressing cells with a reduction in proliferative activity (Hui et al. 2001), and have been used as an in vitro model to study islet neogenesis (Bulotta et al. 2002). We found that the expression of NDRG4A2 and PDX-1 in ARIP cells were increased, while DNA synthesis was inhibited, by incubation with GLP-1. Conversely, prolonged exposure to GLP-1 for 14 days increased the percentage of cells immunoreactive for glucagon, but did not cause the appearance of insulin-positive cells. This suggests that NDRG4 may be up-regulated during commitment to differentiation in the pancreatic duct cells, as happens in neuronal tissues. The use of siRNA to block NDRG4A2 translation resulted in the inhibition of GLP-1 action on DNA synthesis and differentiation, implying that NDRG4 may be a GLP-1-dependent effector gene. N-MYC was also expressed within ARIP cells, and was up-regulated after incubation with GLP-1. N-MYC has been shown to regulate NDRG4 expression in neural tissues, and high levels of N-MYC are expressed during development in the fetal brain, kidney, and within neuroblasts migrating from the neural crest. However, little expression of this gene is found in adult tissues. Targeted expression of human N-MYC selectively caused pancreatic neuroendocrine tumors in transgenic zebra fish (Yang et al. 2004). The expression of both N-MYC and NDRG4 in pancreatic duct cells indicates further similarities in cell differentiation pathways between the pancreas and central nervous system. It is possible that the effects of GLP-1 on NDRG4 in ARIP cells were mediated by an increased expression of N-MYC, but such an indirect pathway cannot be concluded from the present experiments. Exposure of cell clusters derived from primary neonatal rat ductal cultures to GLP-1 caused a small increase in the population of glucagon immunoreactive cells within these clusters, implying that the actions of GLP-1 are not confined to established pancreatic cell lines.
In summary, we have found the expression of NDRG4A2 in rat pancreatic duct cells and in the ARIP pancreatic cell line. The expression of this gene is regulated by GLP-1 and is inversely related to cell proliferation but associated with the expression of PDX-1 and a commitment to cell differentiation. This suggests that NDRG4A2 may play a role in the differentiation of pancreatic cells.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by the Canadian Institutes of Health Research (2005); the Juvenile Diabetes Research Foundation (2003), and the Stem Cell Network Centre of Excellence (2005).
Acknowledgements
We are grateful to Ms Brenda Strutt for expert technical assistance.
References
Alpert S, Hanahan D & Teitelman G 1988 Hybrid insulin genes reveal a developmental lineage for pancreatic endocrine cells and imply a relationship with neurons. Cell 53 295–308.
Bonner-Weir S & Weir GC 2005 New sources of pancreatic β-cells. Nature Biotechnology 23 857–861.
Bonner-Weir S, Toschi E, Inada A, Reitz P, Fonseca SY, Aye T & Sharma A 2004 The pancreatic ductal epithelium serves as a potential pool of progenitor cells. Pediatric Diabetes 2 16–23.
Bulotta A, Hui H, Anastasi E, Bertolotto C, Boros LG, Di Mario U & Perfetti R 2002 Cultured pancreatic ductal cells undergo cell cycle re-distribution and β-cell-like differentiation in response to glucagon-like peptide-1. Journal of Molecular Endocrinology 29 347–360.
Cerf ME 2006 Transcription factors regulating beta-cell function. European Journal of Endocrinology 155 671–679.
Delacour A, Nepote V, Trumpp A & Herrera PL 2004 Nestin expression in pancreatic exocrine cell lineages. Mechanisms of Development 121 3–14.
Frohman MA, Dush MK & Martin GR 1988 Rapid production of full-length cDNAs from rare transcripts: amplification using a single gene specific oligonucleotide primer. PNAS 85 8998–9002.
Gasa R, Mrejen C, Leachman N, Otten M, Barnes M, Wang J, Chakrabarti S, Mirmira R & German M 2004 Proendocrine genes coordinate the pancreatic differentiation program in vitro. PNAS 101 13245–13250.
Githens S 1988 The pancreatic duct cell: proliferative capabilities, specific characteristics, metaplasia, isolation, and culture. Journal of Pediatric Gastroenterology and Nutrition 7 486–506.
Gitlin L, Karelsky S & Andino R 2002 Short interfering RNA confers intracellular antiviral immunity in human cells. Nature 418 430–434.
Habener JF, Kemp DM & Thomas MK 2005 Minireview: transcriptional regulation in pancreatic development. Endocrinology 146 1025–1034.
Hui HX, Wright C & Perfetti R 2001 Glucagon -like peptide induces differentiation of islet duodenal homeobox-1- positive pancreatic ductal cells into insulin-secreting cells. Diabetes 50 785–796.
Joanette E, Reusens B, Arany E, Thyssen S, Remacle RC & Hill DJ 2004 Low protein diet during early life causes a reduction in the frequency of cells immunopositive for nestin and CD34 in both pancreatic ducts and islets in the rat. Endocrinology 145 3004–3013.
Lechner A, Leech CA, Abraham EJ, Nolan AL & Habener JF 2002 Nestin-positive progenitor cells derived from adult human pancreatic islets of Langerhans contain side population (SP) cells defined by expression of the ABCG2 (BCRP1) ATP-binding cassette transporter. Biochemical and Biophysical Research Communications 93 670–674.
Lendhal U, Zimmerman LB & Mckay RD 1990 CNS stem cells express a new class of intermediate filament protein. Cell 60 585–595.
Liu H-K, Green DB, Gault VA, McCluskey JT, McClenaghan NH, O'Harte FP & Flatt PR 2004 N-acetyl-GLP-1: a DPP IV-resistant analogue of glucagon-like peptide-1 (GLP-1) with improved effects on pancreatic β-cell-associated gene expression. Cell Biology International 28 69–73.
Maeda A, Hongo S & Miyazaki A 2004 Genomic organization, expression, and comparative analysis of the non-coding region of the rat Ndrg4 gene. Gene 324 149–158.
Maria-Engler SS, Correa-Giannella ML, Labriola L, Krogh K, Colin C, Lojudice FH, Aito CA, de Oliveira EM, Correa TC & da Silva JC et al. 2004 Co-localization of nestin and insulin and expression of islet cell markers in long-term human pancreatic nestin-positive cell cultures. Journal of Endocrinology 183 455–467.
Murtaugh LC 2007 Pancreas and beta-cell development: from the actual to the possible. Development 134 427–438.
Nakada N, Hongo S, Ohki T, Maeda A & Takeda M 2002 Molecular characterization of NDRG4/Bdm1 protein isoforms that differentially regulated during rat brain development. Developmental Brain Research 135 45–53.
Ohki T, Hongo S, Nakada N, Maeda A & Takeda M 2002 Inhibition of neurite outgrowth by reduced level of NDRG4 protein in antisense transfected PC12 cells. Developmental Brain Research 135 55–63.
Qu X-H, Zhai Y, Wei H-D, Zhang C, Xing G, Yu Y & He F 2002 Characterization and expression of three novel differentiation-related genes belong to the human NDRG gene family. Molecular and Cellular Biochemistry 229 35–44.
Ramiya VK, Maraist M, Arfoors KE, Schatz DA, Peck AR & Cornelius JG 2000 Reversal of insulin-dependent diabetes using islets generated in vitro from pancreatic stem cells. Nature Medicine 6 278–282.
Seaberg RM, Smukler S, Kieffer TJ, Enikolopov G, Asghar Z, Wheeler MB, Korbutt G & van der Kooy D 2004 Clonal identification of multipotent precursors from adult mouse pancreas that generate neural and pancreatic lineages. Nature Biotechnology 22 1115–1124.
Shaw E, McCue LA, Lawrence CE & Dordick JS 2002 Identification of a novel class in the α/β hydrolase fold superfamily: the N-myc differentiation-related proteins. Proteins 47 163–168.
Soria B 2001 In vitro differentiation of pancreatic ß-cells. Differentiation 68 205–219.
Treutelaar MK, Skidmore JM, Dias-Leme CL, Hara M, Zhang L, Simeone D, Martin DM & Burant CF 2003 Nestin lineage cells contribute to the microvasculature but not endocrine cells of the islet. Diabetes 52 2503–2512.
Wang R & Yashpal N 2004 Phenotypic analysis of c-kit expression in epithelial monolayers derived from postnatal rat pancreatic islets. Journal of Endocrinology 182 113–122.
Wang JF, Shen M, Fong GH & Hill DJ 2000 A soluble fibroblast growth factor receptor is released from HL-60 promyelocytic leukemia cells: implication for paracrine growth control. Growth Factors 17 203–214.
Yamauchi Y, Hongo S, Ohashi T, Shioda S, Zhou C, Nakai Y, Nishinaka N, Takahashi R, Takeda F & Takeda M 1999 Molecular cloning and characterization of a novel developmentally regulated gene, Bdm1, showing predominant expression in postnatal rat brain. Molecular Brain Research 68 149–158.
Yang HW, Kutok JF, Lee NH, Piao HY, Fletcher CD, Kanki JP & Look AT 2004 Target expression of human MYCN selectively causes pancreatic neuroendocrine tumors in transgenic zebra fish. Cancer Research 64 7256–7262.
Zhou R-H, Kokame K, Tsukamoto Y, Yutani C, Kato H & Miyata T 2001 Characterization of the human NDRG gene family: a newly identified member, NDRG4, is specifically expressed in brain and heart. Genomics 73 86–97.
