去甲基化和去乙酰化

Biochemical and Biophysical Research Communications 388(2009)418–421

Contents lists available at ScienceDirect

Biochemical and Biophysical Research Communications

journal homepage:www.else v i e r. c o m /l o ca t e /y b b r c

Repression of induced apoptosis in the 2-cell bovine embryo involves DNA methylation and histone deacetylation

Silvia F. Carambula, Lilian J. Oliveira, Peter J. Hansen *

Department of Animal Sciences, University of Florida, P.O. Box 110910, Gainesville, FL 32611-0910, USA

a r t i c l e i n f o a b s t r a c t

Apoptosis in the bovine embryo cannot be induced by activators of the extrinsic apoptosis pathway until the 8–16-cellstage. Depolarization of mitochondria with the decoupling agent carbonyl cyanide 3-chlo-rophenylhydrazone (CCCP)can activate caspase-3in 2-cell embryos but DNA fragmentation does not occur. Here we hypothesized that the repression of apoptosis is caused by methylation of DNA and

TUNEL was affected by a treatment ÂCCCP interaction (P

Article history:

Received 2August 2009

Available online 8August 2009Keywords:Apoptosis

Preimplantation embryo DNA methylation Histone acetylation 5-Aza-20-deoxycytidine Trichostatin A

Introduction

During preimplantation development, the mammalian embryo goes through a period where it is resistant to proapoptotic signals. In the best studied example, the bovine, this period lasts from the 2-cell stage through the 8–16-cellstage [1–5]. Inhibition of the extrinsic pathway for apoptosis at the 2-cell stage is caused in part by resistance of the mitochondria to depolarization [4,5]. In addi-tion, a second block exists that is revealed when the mitochondrial membrane is artiﬁciallydepolarized by carbonyl cyanide 3-chloro-phenylhydrazone (CCCP).In this case, caspase-9and caspase-3activation takes place but DNA fragmentation does not occur [4]. Thus, DNA is resistant to caspase-3mediated events such as activa-tion of caspase-activated DNase (CAD).

One possible explanation for DNA resistance to CAD may reside with the structure of DNA in the early preimplantation embryo. At the 2-cell stage, little transcription takes place [6–7]and DNA is highly methylated [8]. DNA demethylation occurs over the next several cleavage divisions [9]. Thus, the stage of development at which susceptibility to apoptosis is acquired (the8–16-cellstage) is also a time of when DNA methylation is reduced [9]and tran-scription is activated [6].

DNA methylation can reduce the accessibility of DNases to DNA as shown for DNase I [10]. Here we hypothesize that the repression of apoptosis responses in response to mitochondrial depolarization in the 2-cell embryo is caused by DNA methylation that makes internucleosomal DNA inaccessible to activated CAD. Moreover, we hypothesize that repression requires deacetylated histones.

Materials and methods

Reagents. Materials for in vitro maturation of oocytes, in vitro fer-tilization, and embryo culture were obtained as described previously [11]. Carbonyl cyanide 3-chlorophenylhydrazone (CCCP)was pur-chased from Sigma (St.Louis, MO) and was maintained in 100mM stocks in dimethyl sulfoxide (DMSO)at À20°C in the dark. The CCCP stock solution was diluted in embryo culture medium (calledKSOM-BE2, see Ref. [12]for recipe) to 100l M in 0.1%DMSO on the day of use. 5-Aza-20-deoxycytidine (AZA)and trichostatin-A (TSA)were ob-tained from Sigma and used at a ﬁnalconcentration of 100l M and 100nM, respectively. The In Situ Cell Death Detection Kit (TMRred) was from Roche Diagnostics Corporation (Indianapolis,IN), Hoescht 33342was from Sigma, polyvinylpyrrolidone (PVP)was from Eastman Kodak (Rochester,NY). Anti-5-methylcytosine (mouseIgG1; clone 16233D3) was purchased from Calbiochem (SanDiego, CA). The Zenon Alexa Fluor 488mouse IgG1labeling kit 488and Prolong ÒAntifade Kit were obtained from Invitrogen

*Corresponding author. Fax:+[1**********].

E-mail address:[email protected]ﬂ.edu(P.J.Hansen).

S.F. Carambula et al. /Biochemical and Biophysical Research Communications 388(2009)418–421419

Molecular Probes (Eugene,OR). All other reagents were purchased from Sigma or Fisher Scientiﬁc(Pittsburgh,PA).

Experiment 1—Effectsof cytosine demethylation and inhibition of histone deacetylation on induction of apoptosis by CCCP in 2-cell em-bryos. Procedures for production of embryos in vitro were per-formed as previously described [12]. After fertilization of matured oocytes for 8h at 38.5°C in an atmosphere of 5%(v/v)CO 2in humidiﬁedair, putative zygotes were cultured in groups of 30in 50-l l microdrops of KSOM-BE2overlaid with mineral oil at 38.5°C in a humidiﬁedatmosphere of 5%(v/v)CO 2and 5%(v/v) O 2with the balance N 2. At 18h post insemination (hpi),embryos were harvested and placed in groups of 30in fresh 50-l l micro-drops of KSOM-BE2containing either 0.1%DMSO (vehicle),100l M AZA or 100nM TSA. At 28–30hpi, 2-cell embryos were harvested and placed in groups of 10–20in 50-l l microdrops of KSOM-BE2containing the same treatment as previously (vehicle,AZA or TSA) and either vehicle (0.1%DMSO, v/v)or 100l M CCCP. Embryos were cultured for 24h, harvested and then analyzed for TUNEL labeling.

Procedures for TUNEL were performed as described previously [13]. Slides were examined using a Zeiss Axioplan 2epiﬂuores-cence microscope (Zeiss,Gottingen, Germany) with Zeiss ﬁltersets 02(DAPIﬁlter)and 15(rhodamineﬁlter).Digital images for epi-ﬂuorescenceand for light microscopy using differential interfer-ence contrast were acquired using AxioVision software (Zeiss)and a high-resolution black and white Zeiss AxioCam MRm digital camera. Images were merged for presentation. The Hoescht stain-ing was digitally converted to green before merger.

The experiment was replicated six times using a total of 458embryos.

Experiment 2—Effectsof cytosine demethylation and inhibition of histone deacetylation on DNA methylation. The experiment was con-ducted as for Experiment 1except embryos were examined for DNA methylation at the end of the experiment using immunocyto-chemistry with an antibody against 5-methylcytosine. Unless otherwise stated, reactions were at room temperature and re-agents were diluted in phosphate-buffered saline (PBS;10mM KPO 4, pH 7.4containing 0.9%(w/v)NaCl) containing 1mg/mlpol-yvinylpyrrolidone (PVP).Embryos were washed in PBS–PVP,ﬁxedin 4%(w/v)paraformaldehyde, washed in PBS–PVP,permeabilized with 0.3%(v/v)Triton X-100for 30min, washed extensively in 0.05%Tween 20and treated with 3M HCl for 30min at 37°C. After neutralization with 100mM Tris–HCl,pH 8.5containing 1mg/mlPVP, embryos were washed in 0.05%(v/v)Tween 20and non-spe-ciﬁcbinding sites blocked by incubation in a blocking buffer con-sisting of PBS–PVPcontaining 2%(w/v)bovine serum albumin overnight at 4°C.

The anti-5-methylcytosine antibody used for visualization of DNA methylation was labeled with Fab fragments against mouse IgG conjugated to Alexa Flour 488(ZenonÒMouse Labeling IgG kits, Invitrogen Molecular Probes) as per manufacturer’sinstruc-tions. An irrelevant mouse IgG1was similarly labeled as an isotype control. The labeled complex was diluted in blocking buffer at a ﬁ-nal concentration of 5l g/mlprimary antibody and embryos were incubated for 1h at room temperature in the dark. After several washes in 0.05%(v/v)Tween 20in PBS-PVP, embryos were placed on slides and coverslips mounted using Prolong ÒAntifade reagent (Invitrogen).Embryos were examined using a Zeiss Axioplan 2epi-ﬂuorescencemicroscope with Zeiss ﬁltersets 02(DAPIﬁlter)and 03(FITC).Intensity of methylation was subjectively scored for each embryo on a scale of 0(nomethylation) to 3. A total of 61embryos in two replicates were analyzed.

Statistical analysis. Data were analyzed by least-squares analysis of variance using the General Linear Models procedure of the Sta-tistical Analysis System (SASfor Windows, version 9.2, SAS Insti-tute, Inc., Cary NC). Dependent variables for Experiment 1,

calculated on an embryo basis, were total cell number and percent of cells that were apoptotic (i.e.,TUNEL positive). Independent variables included pretreatments (vehicle,AZA or TSA), CCCP (yesvs. no) and replicate. The mathematical model included main ef-fects and all interactions. Replicate was considered random and other main effects were considered ﬁxed.F tests were calculated using error terms calculated from expected means squares. Differ-ences between individual means were determined using the pdiff procedure of SAS. The dependent variable for Experiment 3was methylation score and the independent variable was treatment. Results

Experiment 1—Effectsof cytosine demethylation and inhibition of histone deacetylation on induction of apoptosis by CCCP in 2-cell embryos

In the ﬁrstexperiment, embryos were treated with either AZA or TSA at the zygote stage to block cytosine methylation or histone deacetylation and then treated with CCCP at the 2-cell stage. Rep-resentative images of TUNEL labeling are shown in Fig. 1A–F, least-squares means ±SEM for total cell number are in Fig. 1G and least-squares means ±SEM for the percent of cells that were TUNEL-po-sitive are in Fig. 1H.

Embryo growth, as determined by total cell number at the end of the experiment, was reduced by AZA, and to a lesser extent, TSA (P

As shown in Fig. 1H, the percent of blastomeres positive for TUN-EL was affected by a treatment ÂCCCP interaction (P

The degree of TUNEL labeling in the absence of CCCP was great-er for embryos treated with TSA than for control embryos or em-bryos treated with AZA (P 8cells. Further analysis of the effect of CCCP on TSA-treated embryos focused on the subset of embryos that were

Experiment 2—Effectsof cytosine demethylation and inhibition of histone deacetylation on DNA methylation

As determined by reactivity with an antibody to 5-methylcyto-sine, treatment of putative zygotes with AZA or TSA reduced DNA methylation at 52–54hpi (Fig. 2). In control embryos treated with vehicle, nuclei reacted strongly with antibody against 5-methyl-cytosine (Fig. 2A). Immunoreaction product was greatly reduced in embryos treated with AZA (Fig. 2B) and reduced to a lesser ex-tent for embryos treated with TSA (Fig. 2C). The subjective score for degree of DNA methylation was greatest in control embryos

420S.F. Carambula et al. /Biochemical and Biophysical Research Communications 388(2009)418–421

Fig. 1. DNA demethylation and histone acetylation allows DNA fragmentation in 2-cell embryos after apoptosis triggered by mitochondria depolarization. Putative zygotes were treated with vehicle (DMSO),5-aza-20-deoxycytidine (AZA)or trichostatin-A (TSA);2-cell embryos were collected at 28–30h post insemination and exposed to 100l M CCCP. Total cell number and the percent of cells positive for the TUNEL reaction were determined 24h later. Representative images of embryos following the TUNEL procedure are shown in panels A–F.Nuclei were labeled with Hoechst 33342(digitallycolored as green) and those that are TUNEL-positive nuclei are additionally labeled with TMR red (red).(G)Total cell number and (H)the percent of blastomeres that are TUNEL positive. Data are least-squares means ±SEM for embryos cultured in the absence (blackbars) and presence (whitebars) of CCCP. Cell number was affected by pretreatment (P

less).

(2.5±0.1), least in AZA-treated embryos (1.0±0.1), and intermedi-ate in TSA-treated embryos (1.9±0.1). Differences between each mean were signiﬁcant(P

The bovine preimplantation embryo undergoes a period from the 2-cell stage to 8–16-cellstage when it is resistant to activators of the extrinsic pathway for induction of apoptosis [1–5]. The block to apoptosis involves resistance of the mitochondria to depolariza-tion and failure of caspase-3activation to lead to DNA fragmenta-tion [4,5]. Here we show that the resistance of DNA to caspase-3mediated events is the result of inaccessibility of the DNA caused by a chromatin structure dependent upon DNA methylation and histone acetylation.

5-Aza-20-deoxycytidine inhibits DNA methylation by incor-poration into DNA during replication and subsequent inhibition of DNA methyltransferases (DNMT)[14]. Treatment of embryos with AZA reversed the block to apoptosis so that CCCP treat-ment caused DNA fragmentation. Experiments with AZA indi-cate that inhibition of apoptosis caused by mitochondrial depolarization involves DNA methylation preventing accessibil-ity of CAD to DNA. One can visualize two mechanisms by which methylated cytosines could prevent enzymatic cleavage of DNA. Methylated cytosines can repel certain proteins, for example transcription factors [15], and may also repel CAD. Alternatively, methylated cytosines can attract other proteins, such as the Sin3A histone deacetylase complex and a methyl-CpG binding protein called MeCP2that binds tightly to chro-mosomes [15]

Fig. 2. Reduction in DNA methylation caused by treatment of putative zygotes with 5-aza-20-deoxycytidine (AZA)and trichostatin-A (TSA).DNA methylation was observed by the immunolocalization of 5-methylcytosine in bovine embryos at 52–54h after insemination using an anti-5-methycytosine tagged with Alexa Fluor 488(green).(Forinterpretation of color mentioned in this ﬁgurethe reader is referred to the web version of the article.)

S.F. Carambula et al. /Biochemical and Biophysical Research Communications 388(2009)418–421421

The fact that embryos treated with TSA were also capable of undergoing DNA fragmentation in response to CCCP suggests that repression of apoptosis involves histone interactions with DNA controlled by histone deacetylation. Results with TSA were more complex to interpret than for AZA because more TSA-treated em-bryos not exposed to CCCP experienced TUNEL labeling than for AZA-treated embryos. This effect is due to an increase in TUNEL labeling among TSA-treated embryos that were 8cells or greater. Unlike for AZA, which caused a large reduction in cell number, TSA reduced developmental competence only slightly and many TSA embryos reached the 8–16-cellstage. In these more advanced embryos, TSA caused apoptosis in the absence of CCCP.

Given the role of DNA methylation and histone deacetylation in repressing apoptosis in early stages of development, it is proposed that the acquisition of the capacity for apoptosis at the 8–16-cellstage is dependent upon loss of DNA methylation or changes in his-tone acetylation. There are large species differences in the pattern of DNA methylation during early development with some species like the mouse experiencing a continual reduction in DNA methyl-ation until the blastocyst stage while other species like the pig and rabbit do not experience large scale demethylation during early development [16]. In the cow, DNA methylation is reduced from the 2-cell to 8-cell stage and then increases by the 16-cell stage [9,17]. There are also changes in histone acetylation that occur dur-ing development with Histone H4K5and K12becoming deacety-lated at the one and 2-cell stages, followed by reacetylation that reaches a maximum at the 8-cell stage [17].

Given the importance of DNA methylation and histone deacet-ylation for repressing apoptosis in early cleavage-stage embryos, it is possible that some types of embryonic death result from inad-equate DNA methylation or histone deacetylation. Patterns of DNA methylation during early development are clearly important for embryonic development because AZA caused a large reduction in embryo cell number.

In conclusion, repression of apoptosis in the 2-cell embryo in-volves inaccessibility of caspase-activated DNases to the DNA med-iated by a chromatin structure determined by DNA methylation and histone deacetylation. Future work should focus on the particular interactions between methylated cytosines and histones responsi-ble for this inaccessibility as well as the importance of aberrant chro-matin structure and premature apoptosis in embryonic death. Acknowledgments

Funding was provided by the National Research Initiative Com-petitive Grants Program Grant No. 2007-35203-18070from the U.S. Department of Agriculture Cooperative State Research, Educa-tion and Extension Service. Lilian Oliveira was supported by a CAPES/FulbrightFellowship. The authors thank William Rembert

for collecting ovaries; Marshall, Adam, and Alex Chernin and employees of Central Beef Packing Co. (CenterHill, FL) for provid-ing ovaries; and Scott A. Randell of Southeastern Semen Services (Wellborn,FL) for donating semen. None of the authors have a con-ﬂictof interest. References

[1]F.F. Paula-Lopes, P.J. Hansen, Heat-shock induced apoptosis in preimplantation

bovine embryos is a developmentally-regulated phenomenon, Biol. Reprod. 66(2002)1169–1177.

[2]C.E. Krininger III, S.H. Stephens, P.J. Hansen, Developmental changes in

inhibitory effects of arsenic and heat shock on growth of preimplantation bovine embryos, Mol. Reprod. Dev. 63(2002)335–340.

[3]P. Soto, R.P. Natzke, P.J. Hansen, Actions of tumor necrosis factor-a on oocyte

maturation and embryonic development in cattle, Am. J. Reprod. Immunol. 50(2003)380–388.

[4]A.M. Brad, K.E.M. Hendricks, P.J. Hansen, The block to apoptosis in bovine two-cell embryos involves inhibition of caspase-9activation and caspase-mediated DNA damage, Reproduction 134(2007)789–797.

[5]L.A. de Castro e Paula, P.J. Hansen, Ceramide inhibits development and

cytokinesis and induces apoptosis in preimplantation bovine embryos, Mol. Reprod. Dev. 75(2008)1063–1070.

[6]R.E. Frei, G.A. Schultz, R.B. Church, Qualitative and quantitative changes in

protein synthesis occur at the 8–16-cellstage of embryogenesis in the cow, J. Reprod. Fertil. 86(1989)637–641.

[7]D.R. Natale, G.M. Kidder, M.E. Westhusin, A.J. Watson, Assessment by

differential display-RT-PCR of mRNA transcript transitions and alpha-amanitin sensitivity during bovine preattachment development, Mol. Reprod. Dev. 55(2000)152–163.

[8]J.S. Park, Y.S. Jeong, S.T. Shin, K.K. Lee, Y.K. Kang, Dynamic DNA methylation

reprogramming:active demethylation and immediate remethylation in the male pronucleus of bovine zygotes, Dev. Dyn. 236(2007)2523–2533.

[9]W. Dean, F. Santos, M. Stojkovic, V. Zakhartchenko, J. Walter, E. Wolf, W. Reik,

Conservation of methylation reprogramming in mammalian development:aberrant reprogramming in cloned embryos, Proc. Natl. Acad. Sci. USA 98(2001)13734–13738.

[10]S. Kochanek, D. Renz, W. Doerﬂer,Differences in the accessibility of

methylated and unmethylated DNA to DNase I, Nucleic Acids Res. 21(1993)5843–5845.

[11]K.E. Hendricks, L. Martins, P.J. Hansen, Consequences for the bovine embryo of

being derived from a spermatozoon subjected to post-ejaculatory aging and heat shock:development to the blastocyst stage and sex ratio, J. Reprod. Dev. 55(2009)69–74.

[12]P. Soto, R.P. Natzke, P.J. Hansen, Identiﬁcationof possible mediators of

embryonic mortality caused by mastitis:actions of lipopolysaccharide, prostaglandin F 2a , and the nitric oxide generator, sodium nitroprusside dihydrate, on oocyte maturation and embryonic development in cattle, Am. J. Reprod. Immunol. 50(2003)263–272.

[13]Z. Roth, P.J. Hansen, Involvement of apoptosis in disruption of developmental

competence of bovine oocytes by heat shock during maturation, Biol. Reprod. 71(2004)1898–1906.

[14]W.G. Zhu, G.A. Otterson, The interaction of histone deacetylase inhibitors and

DNA methyltransferases inhibitors in the treatment of human cancer cells, Curr. Med. Chem. Anticancer Agents 3(2003)187–199.

[15]A. Bird, DNA methylation patterns and epigenetic memory, Genes Dev. 16

(2002)6–21.

[16]H. Fulka, J.C. St. John, J. Fulka, P. Hozák,Chromatin in early mammalian

embryos:achieving the pluripotent state, Differentiation 76(2008)3–14.[17]W.E. Maalouf, R. Alberio, K.H. Campbell, Differential acetylation of histone H4

lysine during development of in vitro fertilized, cloned and parthenogenetically activated bovine embryos, Epigenetics 3(2008)199–209.