Proximity ligation in situ assay for monitoring the global DNA methylation in cells
© Hervouet et al; licensee BioMed Central Ltd. 2011
Received: 29 October 2010
Accepted: 6 April 2011
Published: 6 April 2011
Skip to main content
© Hervouet et al; licensee BioMed Central Ltd. 2011
Received: 29 October 2010
Accepted: 6 April 2011
Published: 6 April 2011
DNA methylation has a central role in the epigenetic control of mammalian gene expression, and is required for X inactivation, genomics imprinting and silencing of retrotransposons and repetitive sequences. Thus, several technologies have been developed to measure the degree of DNA methylation.
We here present the development of the detection of protein-protein interactions via the adaptation of the proximity ligation in situ technology to evaluate the DNA methylation status in cells since the quantification of Dnmt1/PCNA interaction in cells reflects the degree of DNA methylation.
This method being directly realizable on cells, it appears that it could suggest a wide range of applications in basic research and drug development. More particularly, this method is specially adapted for the investigations realized from a weak quantity of biologic materiel such as stem cells or primary cultured tumor cells for examples.
DNA methylation has a central role in the epigenetic control of mammalian gene expression, and is required for X inactivation, genomics imprinting and silencing of retrotransposons and repetitive sequences [1, 2]. The loss or the decrease of the 5-methylcytosine number on DNA, i.e. the DNA hypomethylation process, is ones of the earlier hallmarks described in neoplasia . Moreover, several paper reports that the DNA hypomethylation is an oncogenic event leading the chromosomal instability and the oncogenes expression [4, 5]. Literature also mentioned that the degree of DNA hypomethylation has been also associated with the tumor progression and with the prognosis of survival of patients suffering of tumors . Conjugated with the fact that the DNA methylation abnormalities are potentially reversible, all these points encouraged the development of pharmacologic inhibitors of DNA methylation and their use in anti-tumor therapies. Thus, the quantification of degree of DNA methylation in cells appears as a crucial point for its use in diagnosis, prognosis and the evaluation of the response to treatments including DNA methylation modulators.
Several technologies have been developed to measure the degree of DNA methylation . A first group of technologies is focused around the use of methodologies that can quantify the 5-methyl cytosines using reversed-phase high performance liquid chromatography (RP-HPLC), two dimensional thin layer chromatography (2D-TLC), high performance liquid chromatography-mass spectrometry (HPLC-MS), high performance capillary electrophoresis (HPCE) and liquid chromatography-electrospray ionization-tandem mass spectrometry (LC-ESI-MS/MS) [8, 9]. A second group of technologies has been developed to quantify the global 5-methylcytosine via the radiolabelling of CpG sites, ELISA method, pyrosequencing, use of methyl-sensitive enzymes in COBRA method for example [10, 11]. Literature also mentions a third group of technologies in which the degree of DNA methylation is analyzed by measuring the activity of enzymes catalyzing the DNA methylation: the DNA methyltransferases . However, all these methods require multiple steps of works and a sufficient (and significant) quantity of cells initially, and this point is not always possible especially by working with certain primary cultured tumor cells. Thus, and in spite of the fact that the method here exposed is an indirect method of quantification of DNA methylation, it provides the possibility to estimate the DNA methylation status from a weak quantity of biologic materiel such as stem cells or primary cultured tumor cells for examples.
By using ApoTome technology, we can visualize the interactions of endogenous proteins Dnmt1 and PCNA (Figure 1D). Moreover, the use of this method validated the idea that the Dnmt1/PCNA interactions occur in nucleus and permitted to quantify the number of interaction per nucleus.
The degree of DNA methylation can also increase when cells are performing de novo methylation by Dnmt3. To aim this situation, we have used the U87-pORF-Dnmt3a cells since we have already used these cells to promote the methylation of the caspase-8 gene . Under this condition, we observed that the global level of DNA methylation and the number of Dnmt1/PCNA interactions unchanged 2-days after the cells transfection (p = 0.7262 and p = 0.4918) (despite the fact that the methylation level of caspase8 gene was already increased at this point (p < 0.0001)), while we observed an increase of the global level of DNA methylation and of the number of Dnmt1/PCNA interactions 4-days after the cells transfection (p = 0.0292 and p = 0.0142) (Figure 5D). Interestingly, these data suggest that the number of Dnmt1/PCNA interactions increases in response to the Dnmt3a-induced increase of the 5 mC number. However, the Dnmt1/PCNA interactions monitoring was not representative of the methylation degree of the caspase-8 gene. Thus, these data suggest that the Dnmt1/PCNA interactions monitoring is not recommended to analyze the methylation of level of specific gene.
In terms of methodology, all our data argue that the Dnmt1/PCNA Proximity-Ligation In Situ Assay presented here should enable to reflect the degree of DNA methylation in cells. Indeed, that its is following a DNA hypomathylation strategy (inhibition of the Dnmt1 expression or treatment (5aza) or by comparing non tumor cells with tumor cells, the number of Dnmt1/PCNA interactions echoes the degree of DNA methylation. In addition this method being performed from fixated cells, it is adapted for investigation realized from weak quantity of biologic materiel (Stem cells, biopsies samples...) on contrary to the majority of technologies aiming at analyzing the degree of DNA methylation (and mentioned into the introduction section). One other advantage of this method is into the fact that it does no require to transform the cells with plasmid or to extract DNA or nuclear samples in order to perform 5 mC-ELISA, to measure of mMTase, for example. In addition, our data indicate that it is not necessary to take into consideration the expression level of Dnmt1 and/or PCNA since none correlation between the Dnmt1 level expression and the 5 mC number was observed in a large number of glioma and 2) since the Dnmt1/PCNA interactions can be inhibited by other events that the level expression of these proteins (such as the phosphorylation of Dnmt1, the presence of peptide/protein imitating the action of the UP peptide [17, 18].
Supported by these points, we thus envision a wide range of applications for the Dnmt1/PCNA P-LISA in basic research and into the development of DNA hypomethylation strategies based or not on the inhibition of the complexes including the Dnmt1/PCNA interactions such as the one including the Dnmt1/PCNA/UHRF1 which is considered as the main player of the DNA methylation inheritance.
In terms of biologic results, our data show that the breast cancer cell lines (MDA-MD-231 and Cal-51) harbored less Dnmt1/PCNA interactions and less 5-methycytosine (5 mC) than non-tumor breast cells (MCF10A). Thus, breast cancer appears as the second, after glioma, in which we observed that the DNA hypomethylation characterizing the tumor cells is correlated with the decrease of Dnmt1/PCNA interactions [17, 18]. Besides, the fact that DNA of breast cancer cells is hypomethylation in comparison with DNA of breast cells is a well-documented point by the literature [19–21]. The fact that the degree of DNA methylation is correlated with the level of Dnmt1/PCNA interactions makes sense with the fact that the Dnmt1 and PCNA, are two crucial actors, with UHRF1, of the main multiprotein complex, promoting the DNA methylation inheritance in mammalian cells [15, 22]. In addition, our data indicate that the Dnmt1/PCNA interactions are increased, such as the DNA methylation level, during the S phase in comparison with the number of Dnmt1/PCNA interactions and the DNA methylation level characterizing the G0/G1 phase. Thus, the use of the Dnmt1/PCNA P-LISA confirms a dogma of epigenetic reporting that the DNA methylation inheritance catalyzed by the complex including the Dnmt1/PCNA interactions is a process mainly realized during the S phase [15, 22]. In addition, the use of P-LISA could be interesting to monitor the interactions existing between the Dnmt1 protein (the main enzyme catalyzing the DNA methylation inheritance) and different of its interaction partner during the cell cycle. Besides, these investigations are an ongoing subject in our lab and take place into the debate aiming to characterize the dynamism and the kinetic of the DNA methylation inheritance catalyzed by the Dnmt1 during the cell cycle. Our data also indicate that the monitoring of the Dnmt1/PCNA interactions was not representative of the increase of the methylation degree of the caspase-8 gene following the Dnmt3a overexpression, but that at middle term (4-days) the number of Dnmt1/PCNA interactions reflected the Dnmt3a-induced increase of 5 mC. Thus, our data suggest that the monitoring of the Dnmt1/PCNA interactions is not a method adapted to analyze the degree of methylation of a specific gene but is adapted to monitor the degree of DNA methylation. In addition, the increase of Dnmt1/PCNA interactions in Dnmt3a-induced up-methylated cells was not associated with the Dnmt1 or PCNA overexpression (data not shown). Thus, these data suggest that, after a step of de novo methylation catalyzed by the Dnmt3 in "mother cells", "daughter cells" could adapt the number of Dnmt1/PCNA interactions to maintain the DNA methylation pattern realized by the Dnmt3 in "mother cells". Thus, there is an adaptation of the cells according its requirements of DNA methylation inheritance and not according the relative quantity of Dnmt1 and PCNA. This point impacts the methodology of the use of the Dnmt1/PCNA P-LISA for monitoring the global DNA methylation in cells since this suggests that it is not necessary to normalize the counting of P-LISA to the protein level of Dnmt1 and PCNA.
The detection of the Dnmt1/PCNA interactions via the proximity ligation in situ assay to the estimate the degree of DNA methylation in cells appears as an attractive method since 1) it is applicable to a weak quantity of cells, 2) it limits the number of protocol step (no DNA extract, no transfection,...), 3) it reflects the fact that cancer cells are hypomethylated in comparison with no tumor cells, and 4) it echoes the effect of DNA hypomethylating agents/strategies.
In this equation, λ is the emission wavelength, n is the index of medium refraction, ν is the frequency and α is the angle of opening of the objective as previously described . After decovolving (3.5 Huygens Essential software (SVI)), 3D view was obtained by using Amira.4.1.1 program. Finally, the image were analyzed by using the freeware "BlobFinder available for download from http://www.cb.uu.se/~amin/BlobFinder. Thus, we obtained either number of signals per nuclei since nuclei can be automatically identified. In other terms, the use of this program participates to the normalization, standardization, reproducibility and to the definition of the cut off signal to accept/quantify or not a dot.
DNA was extracted by using the QiaAmp DNA mini Kit (Qiagen, France), and 5 mC-ELISA was performed by using the Methylamp Global DNA Methylation Quantification kit (Epigentek-Euromedex, France). DNA ImmunoPrecipitation (MeDIP) were performed by using MethylCollector™ kit according to the manufacturer's instructions (Active Motif, France). qPCR were realized to analyze the level of DNA methylation of a considered gene with specific primers (s: CAGGAGGTGGAGGTTGC andas: GAGCCCTAGACCCTCCC) and according our previous report .
Dnmt1 gene was invalidated by using siRNA approach (sc35204, Tebu-Bio France) according to manufacturer's instructions. siRNA-A was used as control (sc37007, Tebu-Bio France). Classical siRNA treatment (referred as x1) was performed by using 60 pmols of considered siRNA.
This work was supported by a grant from the "Association pour la Recherche contre le Cancer" (ARC#3907 and ARC#1020). We thank to Philippe Juin for giving the breast cell lines.
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.