Differential effects of AdOx on gene expression in P19 embryonal carcinoma cells
© Yan et al; licensee BioMed Central Ltd. 2012
Received: 9 September 2011
Accepted: 6 January 2012
Published: 6 January 2012
Skip to main content
© Yan et al; licensee BioMed Central Ltd. 2012
Received: 9 September 2011
Accepted: 6 January 2012
Published: 6 January 2012
Pluripotent cells maintain a unique gene expression pattern and specific chromatin signature. In this study, we explored the effect of the methyltransferase inhibitor adenosine dialdehyde (AdOx) on pluripotency maintenance and gene expression in P19 embryonal carcinoma cells.
After AdOx treatment, the pluripotency-related gene network became disordered, and the early developmental genes were released from the repression. Remarkably, AdOx caused contrasting effects on the expression of two key pluripotency genes, nanog and oct3/4, with the reduction of the repressive histone marks H3K27me3, H3K9me3 and H3K9me2 only in the nanog gene.
Key pluripotency genes were controlled by different mechanisms, including the differential enrichment of repressive histone methylation marks. These data provided novel clues regarding the critical role of histone methylation in the maintenance of pluripotency and the determination of cell fate in P19 pluripotent cells.
Pluripotent cells, including embryonic stem (ES) cells, embryonic germ cells, and embryonal carcinoma cells, are characterized by their ability to differentiate into all somatic cell types under appropriate conditions . Accordingly, pluripotency-related genes are kept active in pluripotent cells, and early developmental genes are maintained in a 'poised' state (i.e., repressed but ready for activation) . Consistent with the wide array of developmental fates, the chromatin in pluripotent cells is also highly adaptable . The unique gene expression pattern and chromatin signature in pluripotent cells are controlled by a transcription factor network that involves Oct3/4, Sox2, and Nanog . These three core factors are tightly regulated, and even limited fluctuation in their expression may cause significant changes in cell fate [4, 5]. In ES cells, the pluripotency-related genes have relatively high levels of the active histone mark H3K4me3 in their chromatin. In contrast, the chromatin that contains poised developmental genes is associated with a combination of H3K4me3 and repressive H3K27me3 marks . Many histone methyltransferases have been proven to be important for normal embryogenesis . Nevertheless, how histone methylation participates in the maintenance of pluripotency and coordinates the precise expression of the core transcription factors remains largely unknown.
P19 pluripotent cell line is derived from mouse embryonal carcinoma that has the ability to contribute to many normal embryonic tissues after blastocyst infection [7–10]. When exposed to nontoxic concentrations of dimethyl sulfoxide (DMSO) in culture conditions, P19 cells can differentiate into muscle cells. When induced with all trans-retinoic acid (RA), they can be directed into neuronal lineage cells . AdOx indirectly inhibits S-adenosylmethionine (SAM)-dependent methyl-transfer by inhibiting the hydrolysis of the by-product S-adenosylhomocysteine . AdOx has been broadly used for the functional analysis of protein methylation . It has been shown that AdOx treatment during RA induction interferes with the neuronal differentiation of P19 cells .
Here, we show that pre-treatment of AdOx blocks RA-induced P19 cell neuronal differentiation. The impact of AdOx on the expression of pluripotency genes was investigated. We found that in contrast to decreased Oct3/4, the expression of other key pluripotency-related genes is elevated or unchanged in P19 cells treated with AdOx. We then specifically examined the opposing effects of AdOx on the expression of nanog and oct3/4, which were supported by the differential repressive histone methylation of these genes. These results provide an example on the differential control of pluripotency-related genes by histone methylation in P19 cells.
In this study, we demonstrate that AdOx affects the maintenance of pluripotency in P19 cells. Among the pluripotency-related genes, nanog and oct3/4 are differentially regulated by AdOx, and their regulation is correlated with a differential reduction in repressive histone methylation marks in these genes.
The SAM-dependent methyltransferase substrates include lipids, nucleic acids, and a variety of proteins . The effects of AdOx may be universal. After treatment of P19 cells with AdOx, we observed that in addition to the aberrant pluripotent gene expression pattern, AdOx blocked the cell cycle, changed the cell shape and induced expression of the adhesive molecule E-cadherin (data not shown). All of these effects may have contributed to the loss of neuronal potential. In addition, the nonspecific release of early developmental genes may also have disturbed the potential for neuronal fate in P19 cells.
In this study, we examined the expression of six pluripotency-associated genes in P19 cells and their changes in response to AdOx treatment. Utf1 and Lin28 were not responsive to AdOx. In contrast, Sox2, Nanog, and Fgf4 showed higher expression levels following AdOx treatment, whereas the expression of Oct3/4 was reduced. Nanog, Oct3/4 and Sox2 are all critical transcription factors for pluripotency maintenance, and Sox2 functions to maintain the identity of neuronal progenitors, suggesting that the core pluripotency transcription factors perform distinct roles during embryonic development . Recently, it was reported that Sox2 and Oct3/4 have opposite effects in defining mesendoderm and neural ectoderm . We found that Nanog and Oct3/4 were differentially regulated by AdOx, which suggests that they are controlled by different mechanisms. Consistent with this finding, Nanog and Oct3/4 play non-overlapping roles in the establishment of the primitive endoderm and the epiblast . These clues led us to study the histone methylation of these two genes.
To maintain pluripotency, the crucial pluripotency transcription factors are maintained at relatively high but precisely controlled levels [4, 5]. In ES cells, the localized chromatin regions of poised developmental genes have both the active histone mark H3K4me3 and the repressive H3K27me3 mark, whereas the pluripotency-related genes are mainly occupied by relatively high levels of H3K4me3 . However, we found repressive histone methylation marks at pluripotency-related genes at low levels, implying that they potentially have a role in the modulation of pluripotency-related genes. The expression of Nanog in ES cells requires the histone demethylase Jmjd2C to antagonize the repressive H3K9me2 . Our ChIP data revealed that AdOx treatment preferentially causes decreased levels of the three repressive histone methylation marks H3K9me2, H3K9me3, and H3K27me3 on the 5' proximal region of the nanog gene, whereas it has no effects on the oct3/4 gene. In other words, the repressive histone methylations are more involved in the regulation of nanog than in the regulation of oct3/4 in P19 cells. As reported previously, both nanog and oct3/4 are under the control of very similar composite sox-oct elements consisting of neighboring binding sites for Sox2 and Oct3/4 [16, 22]. These observations raised interesting questions: how is the differential regulation of nanog and oct3/4 achieved? Is the underlying mechanism related to functional differences between Nanog and Oct3/4? We expect that further investigations in the fields of ES cells and developing embryos would confirm our results and answer the above questions. Because AdOx may also influence the active histone methylation marks, the methylation of DNA and on non-histone proteins , more specific manipulations on the relevant histone methyltransferases would indeed helpful in elucidating the differential regulation of nanog and oct3/4.
We have shown that the expression of Nanog and Oct3/4 are differentially regulated by AdOx, with the repressive histone marks reduced only on the nanog gene. These data shed light on the critical role of histone methylation in the maintenance of pluripotency and the determination of cell fate in pluripotent cells.
P19 cells were cultured in minimum essential medium (MEM) (Invitrogen) supplemented with 10% (v/v) fetal calf serum. For neuronal differentiation, P19 cells were cultured in medium containing 0.5 μM RA (all trans-retinoic acid, Sigma) for 4 days, and the aggregates were plated as a monolayer and cultured for another 4 days in the absence of RA . AdOx pre-treatment involved adding 20 μM of AdOx (Sigma) to P19 cells for 1 day. Then, AdOx was removed, and the cells were cultured with RA.
The following antibodies were used: monoclonal antibody against Neuronal Class III β-tubulin (Tuj1) from Covance, Inc. (MMS435P); antibody against β-actin (ab8226) from Abcam; antibody against HSP90α from Stressgen (SPS-771); antibodies against GAPDH (MAB347) and Sox2 (AB5603) from Chemicon; antibody against Nanog from BETHYL (A300-397A); antibody against Oct3/4 from Santa Cruz (sc-5279); antibodies against trimethyl-Histone H3 (Lys27, 07-449), trimethyl-Histone H3 (Lys9, 07-442), dimethyl-Histone H3 (Lys9, 07-212), and IgG (12-370) from Upstate Biotech.
Primers used in q-RT-PCR analysis
Whole cell extracts (WCEs) were prepared as described previously . The pellets were washed twice with RIPA buffer and then resuspended in 500 μl RIPA buffer (containing protease inhibitors). After adding SDS to 1%, samples were sonicated for 5 seconds on ice on high; the clear solutions were then stored as "nuclear extracts". The samples were separated in SDS-polyacrylamide gels and analyzed as described previously . The density of bands was analyzed and normalized with internal controls (GAPDH or loading) using AlphaImager™ 2200 (Alpha Innotech Corp).
The cells were first fixed with 2% paraformaldehyde, permeabilized with 0.2% Triton X-100 in PBS for 10 min and then washed and blocked in 1% goat serum. The cells were consecutively incubated in 1:100 diluted mouse anti-Tuj1 antibody at 4°C overnight and the fluorescein isothiocyanate (FITC)-conjugated secondary antibody (1:100) at 37°C for 30 min. The cells were then washed and incubated in DAPI (1:3000) for 10 min followed by two PBS washings. Images were collected using a Nikon Eclipse TE2000-U microscope. DAPI was used to stain the nuclei to count the number of cells. The percentage of cells that were immunoreactive for Tuj1 antigens was determined by capturing images of random fields. DAPI-stained nuclei and cells positive for Tuj1 protein were counted.
The pGL3-nanog-luc (-289/+117) was from Dr. Paul Robson . The pGL-oct3/4-luc (-2143/+30) was from Dr. Duanqing Pei . P19 cells were transfected with VigoFect Reagent (Vigorous Biotech) according to the manufacturer's instructions. pGL3-nanog-luc or pGL-oct3/4-luc reporter plasmids (0.1 μg) and 0.002 μg of control plasmid pRL-TK were co-transfected into the cells. Twenty-four hours after transfection, the medium was replaced with fresh α-MEM (AdOx- or AdOx+). Promoter activity was measured 24 hours later. The activities of both luciferase reporters were determined by the Dual-Luciferase Reporter System (Promega) according to the manufacturer's instructions. The relative luciferase activities, as measured in relative light units, were compared to a co-transfected internal control (pRL-TK). The assay results are shown as the means ± S.D.
Primers used in ChIP-qPCR analysis
Statistical analyses were performed using a two-tailed Student's t-test. All of the data are shown as means with standard deviations. P < 0.05 was considered significant (*), p < 0.01 was considered highly significant (**) and P < 0.001 was considered extremely significant (***).
We thank Dr. Paul Robson for the gift of the pGL3-nanog-luc plasmid. We thank Dr. Duanqing Pei for the gift of the pGL-oct3/4-luc plasmid. This work was supported by the National Natural Science Foundation of China (90919048) and the National Key Scientific Program of China (2011CB964902) and Special Funds of State Key Laboratories (2060204).
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.