Logo of jbcAbout JBCASBMBSubmissionsSubscriptionsContactJBCThis Article
J Biol Chem. Aug 8, 2008; 283(32): 21873–21880.
PMCID: PMC2494936

CCCTC-binding Factor Activates PARP-1 Affecting DNA Methylation Machinery*[S with combining enclosing square]

Abstract

Our previous data have shown that in L929 mouse fibroblasts the control of methylation pattern depends in part on poly(ADP-ribosyl)ation and that ADP-ribose polymers (PARs), both present on poly(ADP-ribosyl)ated PARP-1 and/or protein-free, have an inhibitory effect on Dnmt1 activity. Here we show that transient ectopic overexpression of CCCTC-binding factor (CTCF) induces PAR accumulation, PARP-1, and CTCF poly(ADP-ribosyl)ation in the same mouse fibroblasts. The persistence in time of a high PAR level affects the DNA methylation machinery; the DNA methyltransferase activity is inhibited with consequences for the methylation state of genome, which becomes diffusely hypomethylated affecting centromeric minor satellite and B1 DNA repeats. In vitro data show that CTCF is able to activate PARP-1 automodification even in the absence of nicked DNA. Our new finding that CTCF is able per se to activate PARP-1 automodification in vitro is of great interest as so far a burst of poly(ADP-ribosyl)ated PARP-1 has generally been found following introduction of DNA strand breaks. CTCF is unable to inhibit DNMT1 activity, whereas poly(ADP-ribosyl)ated PARP-1 plays this inhibitory role. These data suggest that CTCF is involved in the cross-talk between poly(ADP-ribosyl)ation and DNA methylation and underscore the importance of a rapid reversal of PARP activity, as DNA methylation pattern is responsible for an important epigenetic code.

Over the past decade our laboratory accumulated evidence that links poly(ADP-ribosyl)ation with DNA methylation. A series of different experimental strategies suggests that blockage of poly(ADP-ribosyl)ation induces in vivo DNA hypermethylation. Previous data showed that inhibition of PARP6 activity introduces an anomalous hypermethylated pattern in genomic DNA (1, 2) and in some CpG island regions (3), suggesting that in the absence of ADP-ribose polymers some DNA regions are no longer protected from methylation. Further experiments showed that PARP activity can also affect the methylation pattern of transfected foreign DNA (4). The combined results of these different experimental approaches allowed us to propose the first method to induce DNA hypermethylation in vivo by treatment of cells in culture with PARP activity inhibitors (5) and to study by atomic force microscopy the effect of the addition of new methyl groups to DNA on chromatin structure in vivo (6).

To provide an explanation for how ADP-ribose polymers control and/or protect DNA methylation patterns, several experimental approaches were used. A mechanism has been suggested in which PARP-1 in its poly(ADP-ribosyl)ated isoform makes DNMT1 catalytically inactive and, thus, inefficient in DNA methylation (7). In this model modified PARP-1 is considered as a molecular adaptor of high negative charge onto which chromatin proteins can be attracted and hosted (8). Several proteins show a greater affinity for ADP-ribose polymers than for DNA (9), so that these polymers compete with DNA for binding of these proteins (10). This noncovalent link, which is very strong (11), is not specifically guided by an attraction between charges; proteins showing high affinity for ADP-ribose polymers have an amino acid domain that is responsible for the interaction with these polymers (12). Deeper analysis demonstrated that the affinity of the noncovalent PAR interactions with specific binding proteins depends on the PAR chain length (13).

Dnmt1 possesses two possible consensus amino acid domains for binding with ADP-ribose polymers and shows a very strong affinity for PARs (7). In vitro experiments have shown that the interaction between DNMT1 and ADP-ribose polymers is stable even in the presence of a 30-fold excess of double strand DNA. Furthermore, the ADP-ribose polymers on automodified PARP-1 almost completely inhibit the catalytic activity of DNMT1, whereas unmodified PARP-1 is not able to inhibit the enzyme. These data, taken together with the facts that PARP-1 and DNMT1 co-immunoprecipitate in vivo and that in this complex PARP-1 is in its modified form (7), suggest that modified PARP-1, trapping DNMT1 through the ADP-ribose polymers, is responsible for the catalytic inactivation of the enzyme in chromatin. This mechanism could be responsible for the protection of the nonmethylated state of CpG islands. In such a scenario, the lack of proper (ADP-ribosyl)ation may lead to new aberrant methylation; in other words, inhibition of PARP activity could allow new methyl groups to be inserted onto DNA. An additional mechanism involved in the functional interplay between PARP and DNA methylation is implied by the observation that inhibition of PARP activity increases both mRNA and protein levels of Dnmt1, the major maintenance methyltransferase, at the G1/S phase border, anomalously increasing the formation of the active complex PCNA-Dnmt1 (14).

To establish whether poly(ADP-ribosyl)ation by itself is involved in the regulation of the promoter region of Dnmt1 gene or of another gene whose product is, in turn, involved in the regulation of Dnmt1 expression, we looked at transcription factors that undergo covalent poly(ADP-ribosyl)ation (15). CTCF, the highly conserved and ubiquitously expressed nuclear factor (16), attracted our attention as this protein, which is one of the major players in imprinting and insulator processes (17), brings together the following two epigenetic events in which we are interested: poly(ADP-ribosyl)ation and DNA methylation. As a chromatin insulator, CTCF links specific consensus sequences located in imprinting control regions. Importantly, it is able to link these regions only if they are unmethylated; furthermore, CTCF binding protects them from de novo methylation (1820). It has recently been shown that CTCF, in the control of imprinting, binds the consensus sequences in its covalently poly(ADP-ribosyl)ated form (21, 22) (the ADP-ribose polymers, present on the N-terminal region of CTCF, increase its molecular mass from 130 to 180 kDa). The functional importance of poly(ADP-ribosyl)ation in the control of imprinting has been shown by experiments in which treatment of cells with 3-aminobenzamide, a competitive inhibitor of PARP activity, affects insulator function of most of the CTCF target sites. Further studies (23) indicate that following ectopic overexpression of full-length CTCF, the 180-kDa CTCF poly(ADP-ribosyl)ated form localizes to the nucleolus; the use of 3-aminobenzamide suggests that translocation of CTCF to the nucleolus is dependent on poly(ADP-ribosyl)ation, and it is the epigenetic event involved in the control of transcription from rDNA in the nucleolus. The N-terminal domain of CTCF, which is the one that undergoes poly(ADP-ribosyl)ation, seems to be necessary in determining the severe inhibition of cell proliferation and clonogenicity observed following ectopic overexpression of CTCF (24).

Our present data provide, for the first time, evidence that CTCF is involved in the cross-talk between poly(ADP-ribosyl)ation and DNA methylation. Here we show by transient ectopic overexpression of CTCF and in vitro experiments the following. (a) PAR accumulation is increased without higher Parp-1 and Parp-2 mRNA and protein expression. (b) PARP-1 and CTCF become poly(ADP-ribosyl)ated. (c) CTCF is capable per se of activating PARP-1 automodification. (d) CTCF interacts with PARP-1 both in vivo and in vitro. (e) DNA methyltransferase activity is inhibited without down-regulation of Dnmt1 mRNA and/or decrease of its nuclear level. (f) Genomic DNA is diffusely hypomethylated, and severe widespread hypomethylation characterizes both centromeric and B1 repetitive DNA sequences. (g) Overexpression of CTCF in PARP-1–/– cells shows direct involvement of PARP-1 in CTCF-induced PAR synthesis. As CTCF does not affect per se the activity of DNMT1, these data altogether confirm the important role played by PARs on DNMT1 activity (7).

EXPERIMENTAL PROCEDURES

Cell Culture, Treatments, and Transfection—L929 mouse fibroblasts and A1 mouse embryonic fibroblasts (PARP-1–/–) were maintained as subconfluent culture in high glucose (4.5 g/liter) Dulbecco's modified Eagle's medium, with 10% fetal calf serum (Cambrex Corp.), 2 mM L-glutamine (Cambrex Corp.), and 50 units/ml Pen-Strep (Cambrex Corp.). Exponentially growing cultures were transfected by Lipofectamine Plus reagent (Invitrogen) adopting the manufacturer's protocol with the following plasmid expression vectors: pCI-CTCF-His tag (kindly provided by Elena M. Klenova, Department of Biological Sciences, University of Essex, UK) for the overexpression of human CTCF, pCI-empty vector as control, and pBabe-PURO for selection of transfectants. To obtain hypomethylated DNA, cells were cultivated for 72 h in standard medium containing 5 μm 5-AZA (Sigma).

Western Blot Analysis—Total cell lysates were obtained by direct lysis of cells in RIPA buffer (50 mm Tris-HCl (pH 8), 150 mm NaCl, 0.5% sodium deoxycholate, 0.1% SDS, 1% Nonidet P-40, 1 mm EDTA). Nuclei were collected from trypsinized and phosphate-buffered saline-washed cells by centrifugation following incubation (30 min) in isolation buffer containing 10 mm Tris-HCl (pH 7.9), 4 mm MgCl2, 1 mm EDTA, 0.5 mm di-thiothreitol, 0.25 mm sucrose, 1% Triton X-100. Nuclear fraction was lysed in RIPA buffer, and protein concentration was determined using the Bradford protein assay reagent (Bio-Rad) with bovine serum albumin (Promega) as standard. Both buffers were supplemented with protease inhibitors (complete EDTA-free, Roche Applied Science). Equal protein amounts were subjected to 6% SDS-PAGE and blotted onto Hybond-ECL nitrocellulose membranes (Amersham Biosciences). The antibodies employed were as follows: mouse monoclonal Ab anti-PARP-1 (C2–10, Alexis Biochemicals), mouse monoclonal Ab anti-PARP-2 (4G8, Alexis), mouse monoclonal Ab anti-PAR (10 HA, Trevigen), mouse monoclonal Ab anti-Dnmt1 (Imgenex), rabbit polyclonal Ab anti-CTCF (Upstate), rabbit polyclonal Ab anti-Sp1 (H-225, Santa Cruz Biotechnology), rabbit polyclonal Ab anti-lamin B1 (AbCam), and goat anti-mouse and anti-rabbit horseradish peroxidase-conjugated antibodies (Santa Cruz Biotechnology).

CTCF Expression in the Baculovirus System—pVLH-CTCF vector, kindly provided by Dr. Klenova (Department of Biological Sciences, University of Essex), was co-transfected with the BD BaculoGold™ Baculovirus DNA (BD Biosciences) into SF9 cells. After the co-transfection, the recombinant baculoviruses were selected and amplified following the manufacturer's instructions. The baculovirus recombinant hCTCF His-tagged protein was purified from infected SF9 cells using the Pro-Bond™ purification system (Invitrogen). bvCTCF was quantified by silver staining (SilverQuest™, Invitrogen) using bovine serum albumin as standard.

In Vitro PARP-1 Activity AssaybvCTCF (0.2, 0.4, or 0.8 pmol) was incubated for 2 h at 30 °C with 4 pmol of human recombinant PARP-1 (Alexis) in 200 μl of poly(ADP-ribosyl)ation buffer containing 50 mm Tris-HCl (pH 8.0), 10 mm MgCl2, 10 mm C2H6OS, and 200 μm NAD, in the presence or absence of 200 ng of DNase I-activated DNA or 20 μm PJ-34 (Sigma). PARP-1 automodification reactions with or without PJ-34 (20 μm) were used as control. Reactions were terminated by 20% trichloroacetic acid precipitation, and ADP-ribose synthesis was assayed by 6% SDS-PAGE followed by Western blotting with anti ADP-ribose antibody. An analogous radiometric assay was performed incubating bvCTCF (0.03, 0.05, 0.1, 0.2, 0.4 pmol) for 2 h at 30 °C with 1 pmol of human recombinant PARP-1 in 50 μl of poly(ADP-ribosyl)ation buffer and in the presence of [3H]NAD (0.3 μCi). Following 20% trichloroacetic acid precipitation, samples were analyzed in a Beckman LS-6800 liquid scintillation spectrometer.

Co-immunoprecipitation (Co-IP)—Nuclei obtained from L929 cells were lysed in IP buffer (50 mm Tris-HCl (pH 7.5), 5 mm EDTA, 300 mm NaCl, 1% Nonidet P-40, 1% Triton X-100) supplemented with protease-inhibitors (complete EDTA-free, Roche Applied Science). Lysates (1.5 mg) were pre-cleared with protein A-(for IP anti CTCF and anti PARP-1) or G (for IP anti Dnmt1)-agarose beads (Upstate) on a rotative shaker at 4 °C for 2 h and 30 min. Pre-cleared lysates were incubated with specific antibodies (rabbit polyclonal Ab anti-CTCF, Upstate; mouse monoclonal Ab anti-Dnmt1, Imgenex; and rabbit polyclonal Ab anti-PARP-1, Alexis) and with normal rabbit or mouse IgG (Santa Cruz Biotechnology) on a rotative shaker at 4 °C. The agarose beads, previously saturated with bovine serum albumin (1 μg/μl) overnight, were added to the lysate/Ab solutions and incubated for 2 h on a rotative shaker at 4 °C. Subsequently, beads were washed in IP buffer and boiled in SDS-PAGE sample buffer, and the eluted proteins were analyzed by SDS-PAGE and Western blotting.

Pulldown Assay—50 μl of ProBond™ resin (Invitrogen) was added to 1.2 ml of 1 × native purification buffer (ProBond™) in the presence or absence of 0.8 pmol of bvCTCF recombinant His-tagged protein. Mix was incubated overnight on a rotative shaker at 4 °C. Then the resin was collected and incubated with 4 pmol of human recombinant PARP-1 (Alexis) in 300 μl of 1 × native purification buffer for 3 h on a rotative shaker at 4 °C. After washing the resin with 1× native purification buffer, proteins were eluted from the beads by boiling them in SDS-PAGE sample buffer. Protein-protein interaction was analyzed by 6% SDS-PAGE and Western blot analysis.

DNA Methyltransferase Activity—Equal amounts of nuclear lysates from cells overexpressing CTCF and relative controls (3–5 μg) were analyzed for the DNA methyltransferase activity by the EpiQuik™ DNA methyltransferase assay kit (Epigentek) following the manufacturer's conditions. This assay was also carried out to check the effect of bvCTCF (0.02, 0.04, and 0.08 pmol) or of same amounts of in vitro poly(ADP-ribosyl)ated PARP-1 (25) on 0.08 pmol of human recombinant DNMT1 (New England Biolabs). Human recombinant PARP-1 (Alexis) (0.08 pmol) was used as control.

RNA Extraction and Reverse Transcription—Total RNA (~1.8 μg), purified by RNeasy mini kit (Qiagen), was subjected to retrotranscription using Superscript first-strand synthesis system (Invitrogen).

Real Time PCR—Expression of mRNA for Parp-1, Parp-2, and Dnmt1 genes was measured by real time PCR using Taq-Man gene expression assays (Applied Biosystems) following the manufacturer's protocol for the absolute standard curve method on iCycler IQ detection system (Bio-Rad). The standard curve was generated using 1:1 serial dilutions (from 100 to 12.5 ng) of cDNA obtained from control at 24 h as reference. PCR efficiency was 90–100% for each set of primers and probe in any experiment. The amplification reaction was performed in duplicate for each sample in 96-well plates. The amount of Parp-1, Parp-2, and Dnmt1 mRNAs was calculated adopting the standard curve method, and normalization was carried out using hypoxanthine-guanine phosphoribosyltransferase (Hprt1) as internal control gene. TaqMan gene expression assay IDs for each set of primers and probe were as follows: Mm00599763m1 (Dnmt1); Mm00500154m1 (Parp-1); Mm00456462m1 (Parp-2); and Mm00446968m1 (Hprt1).

DNA Extraction and Methyl-accepting Ability Assay—DNA was purified from cells using the DNeasy tissue kit (Qiagen), according to the manufacturer's instructions. Methyl-accepting ability assay was carried out in a final volume of 50 μl of 10 mm Tris-HCl (pH 7.9), 10 mm MgCl2, 50 mm NaCl, 1 mm dithiothreitol in the presence of 1 μg of purified DNA and 1 unit of bacterial SssI methylase (New England Biolabs), using as methyl donor 16 μm S-adenosylmethionine plus 10 μCi/ml of [methyl-3H]S-adenosylmethionine (GE Healthcare; specific activity 70–80 Ci/mmol). The reaction was incubated for 1 h at 37 °C and was stopped at 60 °C for 30 min after addition of 1% SDS and 250 μg/ml of proteinase K. The incorporation of labeled methyl groups was evaluated on purified DNA in a Beckman LS-6800 liquid scintillation spectrometer.

Methylation-sensitive Southern Blot Analysis—The DNA methylation level of minor satellite and B1 repeats was evaluated by Southern blot analysis. DNAs (2 μg) were digested with 40 units of MspI or HpaII restriction enzymes for 16 h at 37 °C. After 1.5% agarose gel electrophoresis, the digested DNA was blotted on Hybond-N nylon membrane (Amersham Biosciences), and the presence of new HpaII cutting sites was evidenced by hybridization to 3′-digoxigenin-labeled single strand synthetic oligonucleotides as probes. Labeling of probes and detection was performed using digoxigenin oligonucleotide 3′-end labeling kit and digoxigenin luminescent detection kit (Roche Applied Science). Sequence of probes were as follows: for minor satellite repeats, 5′-GGAAACATGATAAAAACCACAGTGTAGAACATATTAGATGAGTGAGTTACACTGAAAAACACATTCGTTGGAAACGGGATTTGTAGAACAGTGTATATCAATGAGTTACAATGAGAAACATC-3′ (26); for B1 5′-AGTGAGTTCCAGGACAGCCAG-3′ (27). Oligos were made by custom primers synthesis service (Invitrogen). As positive control for DNA demethylation, digestion was performed in parallel on DNA from cells treated for 72 h with 5 μm 5-AZA.

RESULTS

In the first experimental stage, our attention was focused on describing what happens in L929 fibroblast cells following transient or semi-stable ectopic CTCF overexpression (24) (supplemental Fig. S1A). To quantify the proliferative capacity of CTCF-overexpressing cells, population doublings over time were calculated. We found that control cultures doubled in 1 day versus 3 days for the CTCF-overexpressing one. Cell growth diminished by about 50% at 72 h post-transfection (supplemental Fig. S1B); no substantial change in the distribution of the cell population among the different phases of the cell cycle was observed at this time despite the inhibition of cell growth (supplemental Fig. S1E). Clonogenic ability was also affected when analyzed at 48 h post-transfection of pCI-CTCF plus pBabe-PURO (1:20) and puromycin selection for 2 weeks. At this time the clonogenicity was decreased by about 57% (supplemental Fig. S1D). Unlike previously reported data, CTCF ectopic overexpression increased the percentage of dead (supplemental Fig. S1C) and apoptotic cells (supplemental Fig. S1F).

The extent of poly(ADP-ribosyl)ation was analyzed by Western blots of nuclear proteins obtained from adherent cells ectopically overexpressing CTCF and the respective control cells (Fig. 1A). The use of anti-PAR antibodies showed that CTCF overexpression induced PARP activity, leading to poly(ADP-ribosyl)ation of chromatin proteins, including PARP-1. Anti-CTCF antibodies allowed the identification of the poly(ADP-ribosyl)ated form of CTCF as a band running at ~180 kDa (Fig. 1A, top panel); it was impossible to detect this band using the anti-ADP-ribose polymer antibodies because the smearing of polymers, which occurs following PARP-1 automodification, covers the molecular weight region corresponding to the 180 kDa of the modified CTCF (21). PARP-1 poly(ADP-ribosyl)ation increased with CTCF overexpression time with a maximum reached at 48 h. Real time PCR and Western blot analyses showed that this increase was not correlated with greater PARP-1 and PARP-2 mRNA levels or with their higher nuclear levels (Fig. 1, A and B).

FIGURE 1.
In vivo and in vitro experiments addressing the correlation between CTCF and poly(ADP-ribosyl)ation. A, nuclear lysates from cultures at 24, 48, and 72 h post-transfection with pCI-CTCF and pCI were analyzed by SDS-PAGE and Western blot using antibodies ...

To directly test for the effect of CTCF on PARP-1 activity, we used immunometric and radiometric methods. Recombinant PARP-1 was incubated in the presence or absence of different amounts of recombinant bvCTCF under conditions that are typically adopted to induce PARP-1 automodification. The immunometric PARP-1 activity assay (Fig. 1C) was carried out in the presence of human recombinant PARP-1 and DNase I-activated DNA without (lane 1) or with (lane 2) PJ-34 (20 μm), a competitive inhibitor of PARP activity. DNase I-activated DNA is DNA on which some nicks are present, this condition being necessary to induce the poly(ADP-ribosyl)ation of the nick-sensor PARP-1. The reaction was carried out in the presence (Fig. 1C, lanes 4–6) or in the absence (lanes 7–9) of PJ-34, plus addition of increasing amounts of recombinant CTCF (CTCF/PARP-1 molar ratio from 0.05 to 0.2). Finally, PARP activity was assayed following addition of recombinant CTCF (at the same CTCF/PARP-1 molar ratio) but in absence of “activated” DNA (1012). Fig. 1C, lane 3, reports the reaction carried out in the absence of PARP-1, indicating the absence of PARP activity in bvCTCF preparation. As shown, increasing the molar ratio of CTCF/PARP-1 increased the level of poly(ADP-ribosyl)ation of PARP-1. It was surprising that CTCF acted as an activator of PARP-1 modification per se, being able to induce poly(ADP-ribosyl)ation of PARP-1 even in the absence of nicked DNA, which is required to allow PARP-1 activity. The presence of CTCF reduces the effect of the competitive inhibitor PJ-34, meaning that there may be conformational change of PARP-1. In fact, poly(ADP-ribosyl)ation of PARP-1 was not completely inhibited by a PJ-34 concentration of 20 μm. Fig. 1D refers to the radiometric assay carried out in presence of [3H]NAD. The molar ratio CTCT/PARP-1 of 0.4 increases PARP-1 activity in presence of activated DNA by about 15-fold. This assay confirms the finding that CTCF is capable per se of inducing poly(ADP-ribosyl)ation of PARP-1 even in absence of activated DNA. In fact, the presence of CTCF at 0.4 pmol allows ~8.5-fold increase in the modification of PARP-1. Co-immunoprecipitation and pulldown experiments have shown that CTCF and PARP-1 directly interact both in vivo and in vitro. (Fig. 1, E and F).

Our knowledge that DNA methylation and poly(ADP-ribosyl)ation are connected prompted us to further investigate the scenario where the ectopic overexpression of CTCF induced a burst of PARs and poly(ADP-ribosyl)ation of PARP-1, which remained high in time. We expected that this condition would induce permanent inhibition of Dnmt1 and consequently genomic DNA hypomethylation. We thus tested whether the increased amount of modified PARP-1 influenced the endogenous DNA methyltransferase activity. The activity assay, carried out on nuclear extracts obtained from cells ectopically overexpressing CTCF and the respective control cells, showed that DNA methyltransferase activity decreased by about 65–70% (Fig. 2C). Real time PCR (Fig. 2B) and Western blot analyses (Fig. 2A) excluded that inhibition of enzymatic activity was dependent on down-regulation of Dnmt1 levels. The possibility that CTCF per se was acting as an inhibitor of DNMT1 activity was disproved by in vitro assays; parallel experiments carried out with in vitro poly(ADP-ribosyl)ated PARP-1 confirmed that poly(ADP-ribosyl)ated PARP-1 is responsible for inhibitory effect on DNMT1 (7) (Fig. 2D). Co-immunoprecipitation experiments, carried out on nuclear extracts from L929 mouse fibroblasts, indicate that PARP-1 and Dnmt1 are associated in vivo (Fig. 2E), confirming our previous data (7).

FIGURE 2.
In vivo and in vitro experiments addressing the effect of CTCF overexpression on DNA methyltransferase activity. A, nuclear lysates from cultures at 24, 48, and 72 h post-transfection with pCI-CTCF and pCI were analyzed by SDS-PAGE and Western blot using ...

Research was then focused on DNA methylation pattern and the possibility that inhibition of Dnmt1 activity leads to DNA hypomethylation. Fig. 3A reports data from methyl-accepting ability assays, carried out on DNA purified from cells overexpressing CTCF and control cells in the presence of labeled S-adenosylmethionine and human recombinant DNMT1. CTCF overexpression introduced hypomethylation on genomic DNAs, and as a result, its methyl-accepting ability increased 3-fold at 24 h from transfection. The integrity and the amounts of purified genomic DNA were evaluated by agarose gel electrophoresis (data not shown). Southern blot analyses were performed on the same DNA samples to localize the new hypomethylated sequences. Fig. 3, B and C, reports experiments in which genomic DNA, previously digested with methylation-sensitive HpaII restriction enzyme and with its insensitive isoschizomer MspI, was analyzed, after electrophoretic separation of DNA fragments, by using 3′-digoxigenin-labeled single strand synthetic oligonucleotides as probes. As can be seen, both centromeric and B1 repetitive DNA sequences were hypomethylated. To verify how specific the PARP-1 poly(ADP-ribosyl)ation is in affecting DNA methylation activity, we performed experiments of CTCF overexpression in PARP-1–/– mouse cells. Ectopic overexpression of CTCF in PARP-1–/– cells did not reduce the DNA methyltransferase activity, Dnmt1 level, and the cell growth significantly (Fig. 4, A–C and supplemental Experimental Procedures), although it induces a moderate increase of PARs because of an enhanced activity of PARP-2 and/or of other members of PARP family (Fig. 4A).

FIGURE 3.
Effect of DNA methyltransferase activity inhibition on the methylation state of genomic DNA. A, methyl-accepting ability assay was carried out on genomic DNA purified from cells transfected for different times (24, 48, and 72 h) with pCI-CTCF (gray bar ...
FIGURE 4.
Effect of CTCF overexpression in PARP-1–/– cells on PARs level, cell survival, Dnmt1 expression, and DNA methyltransferase activity. A, nuclear lysates from L929 and A1 (PARP-1–/–) cells at 72 h post-transfection with pCI ...

DISCUSSION

This research emphasizes the importance of proper dynamics between poly(ADP-ribose) polymerase and poly(ADP-ribose) glycohydrolase (PARG) activities so that cellular homeostasis is not affected by low or excessively high PAR levels. In physiological conditions, the active, poly(ADP-ribosyl)ated PARP-1 molecules constitute a small proportion of the total population of PARP-1 molecules, whereas upon injury to the cell, the number of active PARP-1 molecules increases causing an up to 500 times higher level of PARs (28). In this environment PARP-1 plays an important role as it is responsible for about 90% of PAR production (29). However, the basal level of polymers is rapidly restored thanks to the intervention of PARG (3032), which shuttles from the cytoplasm to the nucleus (33, 34), digesting the excess of nuclear PARs in a few minutes (28, 35).

Over the years, it has been shown that the decrease and increase of PAR level can both threaten the normal cellular physiology. Treatment of cells with competitive inhibitors of PARP activity has demonstrated that delay in DNA damage repair and genomic instability occur when nuclear levels of PARs are low (36). Research on cells knocked out for PARP-1 (37) or PARP-2 (38), the two best-known members of the PARP family (39), has indicated that both enzymes are involved in the mechanism of DNA damage repair: knocking out of either enzyme is not lethal, whereas the cell cannot survive the double knock-outs (38). Likewise harmful is the accumulation of PARs, which is even perceived as a signal for cell death (4042). Cell death, observed when PAR level is high, has long been attributed to NAD+ depletion and thus to energy insufficiency (43). Without excluding that more than one event leads to cellular death, it has been shown that PARs by themselves cause cell mortality (44). Cytoplasmic long and branched polymers (41), interacting with mitochondrial membrane stimulate the release of the pro-apoptotic factor AIF, an apoptotic signal downstream of PARP activation (42, 45, 46).

In the above-mentioned research attention was focused on what happens to the cell when high levels of PARs are present in the cytosol. We have focused our attention on events that occur in the nucleus. Based on our experimental data, we suggest a model (Fig. 5) that depicts the order in which molecular events take place. The starting point is when the increased nuclear level of CTCF induces a higher level of nuclear PARs. It is surprising that at all observation times, 24, 48, and 72 h following ectopic transfection of CTCF, the nuclear level of PARs remained high not depending on an increased PARP-1 expression, even though the PAR level returns to basal level in a few minutes following stress (28).

FIGURE 5.
A possible model describing the sequence of events occurring in cells following CTCF ectopic overexpression. CTCF activates (gray arrow) PARP-1 inducing its automodification and poly(ADP-ribosyl)ation of CTCF and of other nuclear proteins yet to be identified. ...

In vitro experiments show that CTCF is by itself capable of activating PARP-1. This activation happens even in absence of activated DNA, and this is different from YY1 (47) and VP1 (48). Furthermore, the effect of the PJ-34 inhibitor is diminished in presence of CTCF. Co-immunoprecipitation and pulldown experiments confirmed interaction between CTCF and PARP-1 both in vivo and in vitro.

Our previous data show that both free PARs and PARs present on modified PARP-1, through a noncovalent interaction with DNMT1, inhibit its enzymatic activity. Here we investigated whether prolonged elevated PAR levels could stably inhibit the activity of Dnmt1. We show that DNA methyltransferase activity is decreased by about 70% in nuclear extract in cells overexpressing CTCF versus cells transfected with empty vector. A more thorough examination of Dnmt1 has shown that the inhibition is not because of a decreased level of Dnmt1 expression. An in vitro approach allowed us to exclude direct inhibition of DNMT1 activity by CTCF, in line with our previous observation that PARs present on modified PARP-1 inhibit DNMT1 activity (7). The persistence of a high PAR level affects the DNA methylation machinery; Dnmt1 activity is inhibited, leading to wide hypomethylation of the genome, involving both centromeric and B1 repeats. Ectopic overexpression of CTCF in PARP-1–/– cells demonstrates a specific functional relationship between CTCF and PARP-1.

This finding underscores the importance of a rapid reversal of PARP-1 automodification because is affects DNA methylation patterns considered to be part of the epigenetic code. DNA hypomethylation could decondense chromatin structure (49, 50), cause genomic instability (51), direct cells toward apoptosis (52), and cancer (53, 54). The unusual cellular environment, in which a large amount of ADP-ribose polymers are present for several days, mimics the situation in PARG knock-outs and provides a model for the study of damage because of excess of ADP-ribose polymers in cells. Our data suggest that deregulation of DNA methylation pattern is one of the molecular mechanisms causing lethality in PARG knock-out mice (40).

The finding that high PAR levels induce hypomethylation is intriguing, because in tumor cells the hypomethylation of repeat-rich heterochromatin contributes to genomic instability, through increased mitotic recombination events (51). The introduction of new methyl groups onto DNA or the diffuse hypomethylation in cancer cells (54, 56) could occur through deregulation of PARP or PARG activities. This demethylation is a passive mechanism (55) that does not depend on decreased expression of Dnmt1 but on the absence of an active Dnmt1.

In the cellular scenario dependent on ectopic overexpression of CTCF, the cytotoxicity is accounted for by a decreased level of intracellular NAD+ (43), whereas the increased number of apoptotic cells can be explained by recent data showing that the higher ADP-ribose polymer production (57) and DNA hypomethylation (52) induce apoptosis. The connection between DNA methylation and poly(ADP-ribosyl)ation inferred from our previous studies (16) is strengthened by our new findings showing that activation of PARP-1, dependent on CTCF, affects DNA methylation machinery. It is clear that, because of the complexity of pathways that occur in cells, despite evidence showing the connection between poly-ADP-ribosylated PARP-1 and Dnmt1, other additional and/or alternative mechanism(s) cannot be excluded. Based on our extensive research on poly(ADP-ribosyl)ation and DNA methylation, we think that the control of DNA methylation pattern is among the multiple housekeeping roles of PARP-1; this role could be played by the inhibitory effect exerted by PARs on the Dnmt1 (7). The balance of PARs constitutively present in cells may be extremely important; their decrease causes anomalous DNA hypermethylation (16), whereas their increase leads to widespread DNA hypomethylation.

Further research will be carried out to define in depth the role played by CTCF in the cross-talk between poly(ADP-ribosyl)ation and DNA methylation and to establish whether and how the normal functions of PARG are affected in this new cellular scenario. We will assess whether CTCF represses PARG expression, inhibits its enzymatic activity, or impairs its physiological shuttling from the cytoplasm to the nucleus.

Supplementary Material

[Supplemental Data]

Notes

Author's Choice—Final version full access.

*The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

[S with combining enclosing square]The on-line version of this article (available at http://www.jbc.org) contains supplemental Experimental Procedures and Fig. S1.

Footnotes

6The abbreviations used are: PARP, poly(ADP-ribose) polymerase; CTCF, CCCTC-binding factor; PAR, poly(ADP-ribose); Dnmt1, DNA methyltransferase 1; 5-AZA, 5-azacytidine; IP, immunoprecipitation; Co-IP, co-immunoprecipitation; Ab, antibody; PARG, poly(ADP-ribose) glycohydrolase.

References

1. Zardo, G., D'Erme, M., Reale, A., Strom, R., Perilli, M., and Caiafa, P. (1997) Biochemistry 36 7937–7943 [PubMed]
2. De Capoa, A., Febbo, F. R., Giovannelli, F., Niveleau, A., Zardo, G., Marenzi, S., and Caiafa, P. (1999) FASEB J. 13 89–93 [PubMed]
3. Zardo, G., and Caiafa, P. (1998) J. Biol. Chem. 273 16517–16520 [PubMed]
4. Zardo, G., Marenzi, S., Perilli, M., and Caiafa, P. (1999) FASEB J. 13 1518–1522 [PubMed]
5. Zardo, G., Reale, A., Perilli, M., de Capoa, A., and Caiafa, P. (1999) Gene Ther. Mol. Biol. 4 291–296
6. Karymov, M. A., Tomschik, M., Leuba, S. H., Caiafa, P., and Zlatanova, J. (2001) FASEB J. 15 2631–2641 [PubMed]
7. Reale, A., De Matteis, G., Galleazzi, G., Zampieri, M., and Caiafa, P. (2005) Oncogene 24 13–19 [PubMed]
8. Althaus, F. R. (1992) J. Cell Sci. 102 663–670 [PubMed]
9. Malanga, M., and Althaus, F. R. (2005) Biochem. Cell Biol. 83 354–364 [PubMed]
10. Realini, C. A., and Althaus, F. R. (1992) J. Biol. Chem. 267 18858–18865 [PubMed]
11. Panzeter, P. L., Realini, C. A., and Althaus, F. R. (1992) Biochemistry 31 1379–1385 [PubMed]
12. Pleschke, J. M., Kleczkowska, H. E., Strohm, M., and Althaus, F. E. (2000) J. Biol. Chem. 275 40974–40980 [PubMed]
13. Fahrer, J., Kranaster, R., Altmeyer, M., Marx, A., and Burkle, A. (2007) Nucleic Acids Res. 35 e143. [PMC free article] [PubMed]
14. Zardo, G., Reale, A., Passananti, C., Pradhan, S., Buontempo, S., De Matteis, G., Adams, R. L. P., and Caiafa, P. (2002) FASEB J. 16 1319–1321 [PubMed]
15. Hassa, P. O., Haenni, S. S., Elser, M., and Hottiger, M. O. (2006) Microbiol. Mol. Biol. Rev. 70 789–829 [PMC free article] [PubMed]
16. Ohlsson, R., Renkawitz, R., and Lobanenkov, V. (2001) Trends Genet. 17 520–527 [PubMed]
17. Wallace, J. A., and Felsenfeld, G. (2007) Curr. Opin. Genet. Dev. 17 400–407 [PMC free article] [PubMed]
18. Bell, A. C., and Felsenfeld, G. (2000) Nature 405 482–485 [PubMed]
19. Hark, A. T., Schoenherr, C. J., Katz, D. J., Ingram, R. S., Levorse, J. M., and Tilghman, S. M. (2000) Nature 405 486–489 [PubMed]
20. Bell, A. C., West, A. G., and Felsenfeld, G. (2001) Science 291 447–450 [PubMed]
21. Yu, W. Q., Ginjala, V., Pant, V., Chernukhin, I., Whitehead, J., Docquier, F., Farrar, D., Tavoosidana, G., Mukhopadhyay, R., Kanduri, C., Oshimura, M., Feinberg, A. P., Lobanenkov, V., Klenova, E., and Ohlsson, R. (2004) Nat. Genet. 36 1105–1110 [PubMed]
22. Klenova, E., and Ohlsson, R. (2005) Cell Cycle 4 96–101 [PubMed]
23. Torrano, V., Navascues, J., Docquier, F., Zhang, R., Burke, L. J., Chernukhin, I., Farrar, D., Leon, J., Berciano, M. T., Renkawitz, R., Klenova, E., Lafarga, M., and Delgado, M. D. (2006) J. Cell Sci. 119 1746–1759 [PubMed]
24. Rasko, J. E. J., Klenova, E. M., Leon, J., Filippova, G. N., Loukinov, D. I., Vatolin, S., Robinson, A. F., Hu, Y. J., Ulmer, J., Ward, M. D., Pugacheva, E. M., Neiman, P. E., Morse, H. C., Collins, S. J., and Lobanenkov, V. V. (2001) Cancer Res. 61 6002–6007 [PubMed]
25. Blenn, C., Althaus, F. R., and Malanga, M. (2006) Biochem. J. 396 419–429 [PMC free article] [PubMed]
26. Bouzinba-Segard, H., Guais, A., and Francastel, C. (2006) Proc. Natl. Acad. Sci. U. S. A. 103 8709–8714 [PMC free article] [PubMed]
27. Schmid, C., Heng, H. H. Q., Rubin, C., Ye, C. J., and Krawetz, S. A. (2001) Mol. Hum. Reprod. 7 903–911 [PubMed]
28. D'Amours, D., Desnoyers, S., D'Silva, I., and Poirier, G. G. (1999) Biochem. J. 342 249–268 [PMC free article] [PubMed]
29. Shieh, W. M., Ame, J. C., Wilson, M. V., Wang, Z. Q., Koh, D. W., Jacobson, M. K., and Jacobson, E. L. (1998) J. Biol. Chem. 273 30069–30072 [PubMed]
30. Braun, S. A., Panzeter, P. L., Collinge, M. A., and Althaus, F. R. (1994) Eur. J. Biochem. 220 369–375 [PubMed]
31. Lin, W. S., Ame, J. C., AboulEla, N., Jacobson, E. L., and Jacobson, M. K. (1997) J. Biol. Chem. 272 11895–11901 [PubMed]
32. Davidovic, L., Vodenicharov, M., Affar, E. B., and Poirier, G. G. (2001) Exp. Cell Res. 268 7–13 [PubMed]
33. Bonicalzi, M. E., Vodenicharov, M., Coulombe, M., Gagne, J. P., and Poirier, G. G. (2003) Biol. Cell 95 635–644 [PubMed]
34. Ohashi, S., Kanai, M., Hanai, S., Uchiumi, F., Maruta, H., Tanuma, S., and Miwa, M. (2003) Biochem. Biophys. Res. Commun. 307 915–921 [PubMed]
35. Tulin, A., and Spradling, A. (2003) Science 299 560–562 [PubMed]
36. Beneke, S., and Burkle, A. (2007) Nucleic Acids Res. 35 7456–7465 [PMC free article] [PubMed]
37. Wang, Z. Q., Auer, B., Stingl, L., Berghammer, H., Haidacher, D., Schweiger, M., and Wagner, E. F. (1995) Genes Dev. 9 509–520 [PubMed]
38. Ménissier de Murcia, J., Ricoul, M., Tartier, L., Niedergang, C., Huber, A., Dantzer, F., Schreiber, V., Amé, J. C., Dierich, A., LeMeur, M., Sabatier, L., Chambon, P., and de Murcia, G. (2003) EMBO J. 22 2255–2263 [PMC free article] [PubMed]
39. Ame, J. C., Spenlehauer, C., and de Murcia, G. (2004) BioEssays 26 882–893 [PubMed]
40. Koh, D. W., Lawler, A. M., Poitras, M. F., Sasaki, M., Wattler, S., Nehls, M. C., Stoger, T., Poirier, G. G., Dawson, V. L., and Dawson, T. M. (2004) Proc. Natl. Acad. Sci. U. S. A. 101 17699–17704 [PMC free article] [PubMed]
41. Andrabi, S. A., Kim, N. S., Yu, S. W., Wang, H., Koh, D. W., Sasaki, M., Klaus, J. A., Otsuka, T., Zhang, Z., Koehler, R. C., Hurn, P. D., Poirier, G. G., Dawson, V. L., and Dawson, T. M. (2006) Proc. Natl. Acad. Sci. U. S. A. 103 18308–18313 [PMC free article] [PubMed]
42. Yu, S. W., Andrabi, S. A., Wang, H., Kim, N. S., Poirier, G. G., Dawson, T. M., and Dawson, V. L. (2006) Proc. Natl. Acad. Sci. U. S. A. 103 18314–18319 [PMC free article] [PubMed]
43. Chiarugi, A. (2002) Trends Pharmacol. Sci. 23 122–129 [PubMed]
44. Heeres, J. T., and Hergenrother, P. J. (2007) Curr. Opin. Chem. Biol. 11 644–653 [PubMed]
45. Yu, S. W., Wang, H. M., Poitras, M. F., Coombs, C., Bowers, W. J., Federoff, H. J., Poirier, G. G., Dawson, T. M., and Dawson, V. L. (2002) Science 297 259–263 [PubMed]
46. Koh, D. W., Dawson, T. M., and Dawson, V. L. (2005) Pharmacol. Res. 52 5–14 [PubMed]
47. Griesenbeck, J., Ziegler, M., Tomilin, N., Schweiger, M., and Oei, S. L. (1999) FEBS Lett. 443 20–24 [PubMed]
48. Carbone, M., Reale, A., Di Sauro, A., Sthandier, O., Garcia, M. I., Maione, R., Caiafa, P., and Amati, P. (2006) J. Mol. Biol. 363 473–485
49. Bird, A. P., and Wolffe, A. P. (1999) Cell 99 451–454 [PubMed]
50. Wolffe, A. P., and Matzke, M. A. (1999) Science 286 481–486 [PubMed]
51. Eden, A., Gaudet, F., Waghmare, A., and Jaenisch, R. (2003) Science 300 455–455 [PubMed]
52. Jackson-Grusby, L., Beard, C., Possemato, R., Tudor, M., Fambrough, D., Csankovszki, G., Dausman, T., Lee, P., Wilson, C., Lander, E., and Jaenisch, R. (2001) Nat. Genet. 27 31–39 [PubMed]
53. Gaudet, F., Hodgson, J. G., Eden, A., Jackson-Grusby, L., Dausman, J., Gray, J. W., Leonhardt, H., and Jaenisch, R. (2003) Science 300 489–492 [PubMed]
54. Robertson, K. D. (2005) Nat. Rev. Genet. 6 597–610 [PubMed]
55. D'Alessio, A. C., and Szyf, M. (2006) Biochem. Cell Biol. 84 463–476 [PubMed]
56. Jones, P. A., and Baylin, S. B. (2002) Nat. Rev. Genet 3 415–428 [PubMed]
57. Ethier, C., Labelle, Y., and Poirier, G. G. (2007) Apoptosis 12 2037–2049 [PubMed]

Articles from The Journal of Biological Chemistry are provided here courtesy of American Society for Biochemistry and Molecular Biology

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...