Author's address (for correspondence): MC Gomez, Audubon Center for Research of Endangered Species, 14001 River Road, New Orleans, LA 70131, USA. E-mail: firstname.lastname@example.org
Somatic cell nuclear transfer offers the possibility of preserving endangered species including the black-footed cat, which is threatened with extinction. The effectiveness and efficiency of somatic cell nuclear transfer (SCNT) depends on a variety of factors, but ‘inappropriate epigenetic reprogramming of the transplanted nucleus is the primary cause of the developmental failure of cloned embryos. Abnormal epigenetic events such as DNA methylation and histone modifications during SCNT perturb the expression of imprinted and pluripotent-related genes that, consequently, may result in foetal and neonatal abnormalities. We have demonstrated that pregnancies can be established after transfer of black-footed cat cloned embryos into domestic cat recipients, but none of the implanted embryos developed to term and the foetal failure has been associated to aberrant reprogramming in cloned embryos. There is growing evidence that modifying the epigenetic pattern of the chromatin template of both donor cells and reconstructed embryos with a combination of inhibitors of histone deacetylases and DNA methyltransferases results in enhanced gene reactivation and improved in vitro and in vivo developmental competence. Epigenetic modifications of the chromatin template of black-footed cat donor cells and reconstructed embryos with epigenetic-modifying compounds enhanced in vitro development, and regulated the expression of pluripotent genes, but these epigenetic modifications did not improve in vivo developmental competence.
The application of interspecies somatic cell nuclear transfer (SCNT) for the conservation of the black-footed cat (BFC; Felis nigripes), which is threatened with extinction, has been hindered by the low cloning efficiency. We have previously reported that 45–57% of domestic cat (DSH; Felis silvestris catus) recipients became pregnant after the transfer of interspecies BFC (iBFC) -cloned embryos, whereby 3–4% of the transplanted embryos implanted (Gómez et al. 2011). However, none of these implantations developed to full-term, as opposed to the survival birth-rate for interspecies African wildcat (1%; Gómez et al. 2004), sand cat (0.9%; Gómez et al. 2008) and intra-species domestic cat clones (3.3%; Gómez et al. 2011). The effectiveness and efficiency of SCNT depends on a variety of factors, but ‘inappropriate epigenetic reprogramming’ of the transplanted nucleus is the primary cause of the developmental failure of cloned embryos. In iBFC-cloned embryos, aberrant remodelling of the covalent pattern of dimethylation in H3K9 as well as the abnormal or non-expression of pluripotent genes were associated to foetal losses (Gómez et al. 2011).
Methods to overcome abnormal nuclear reprogramming have been addressed using ‘epigenetic-modifying compounds’. In fact, partial erasure of epigenetic marks in bovine fibroblasts treated with either a DNA methyltransferase inhibitor (5-aza-2′deoxycytidine; 5-aza-dC) or an inhibitor of histone deacetylases activity trichostatin A (TSA) improved the in vitro development of cloned embryos (Enright et al. 2003). However, continuously treating both cells and cloned embryos with a combination of these two epigenetic-modifying compounds resulted in a significant increase in blastocyst rate, embryo quality and survival rate to birth (Ding et al. 2008; Wang et al. 2011a,b). In the present study, we evaluated whether treating BFC fibroblasts before SCNT or continuously treatment of iBFC-cloned embryos with epigenetic-modifying compounds could facilitate nuclear reprogramming and improve in vitro and in vivo developmental competence.
Materials and Methods
Domestic cats used as oocyte donors and embryo recipients, and the BFC male were housed at the Audubon Center for Research of Endangered Species (ACRES).All animal procedures were IACUC approved.
All chemicals were obtained from Sigma-Aldrich Chemical Co. (St Louis, MO, USA) unless otherwise stated.
Establishment and culture of donor fibroblasts
Black-footed cat fibroblasts were generated from skin tissue of one adult BFC male (Gómez et al. 2011). Fibroblasts frozen at passages 2 were thawed, and cultured until cells reached 100% confluence and used immediately for analysis or SCNT.
Covalent modifications of H3K9ac and DNA methylation in BFC fibroblasts
The relative levels of H3K9ac and global DNA methylation in BFC fibroblasts treated with: (i) 50 nm TSA or 500 nm SAID for 20 h, (ii) 5-aza-dC (1 nm or 10 nm) for either 20 h or 72 h or (iii) untreated (control) were assessed by flow cytometry (FCM). Cells from each treatment were incubated with primary rabbit anti-acetyl-Histone H3 (lys9) (1 : 100; H3K9ac; Cat# 07-352; Upstate Biotechnology/Millipore, Billerica, MA, USA) or mouse anti-5-methylcytosine (1 : 1000; 5-MeC; Cat# NA81; Calbiochem/Millipore) antibodies and secondary goat anti-rabbit-FITC conjugated (1 : 50; Cat# AP307F, Chemicon/Millipore) or sheep anti-mouse-FITC conjugated (1 : 50; Cat # F2266; Sigma-Aldrich, St. Louis, MO, USA) antibodies, respectively. H3K9ac levels were assessed by FCM following a protocol previously reported (Gómez et al. 2008), while 5-MeC was assessed using a protocol reported by Ronzoni et al. (2005). The average profiles represented as the ratio of H3K9ac/DNA or 5-MeC/DNA signals were used to determine differences between treatments.
For determining the percentage of cells undergoing apoptosis, BFC fibroblasts that were treated with TSA, SAID or 5-aza-dC were cultured in Caspase-3 solution (5 um; NucView™, Biotium Inc. Hayward, CA, USA) following a protocol previously reported (Gómez et al. 2008). Samples of 20 000 cells each in triplicate were analysed by FCM and quadrant analysis detected. Unstained cells were classified as live, and cells that displayed green fluorescence only (live apoptotic) or red and green fluorescence (dead apoptotic) were classified as apoptotic cells.
Embryo production by SCNT and transfer into recipient domestic cats
In vivo-matured oocytes were collected by laparoscopic aspiration of pre-ovulatory ovarian follicles as previously described and SCNT conducted according to methods previously reported by Gómez et al. (2008). Cloned embryos were cultured: (i) for 24 h before fixation, (ii) for 18–36 h before being transferred to recipients or (iii) cultured in vitro until day 8 at which time development to the blastocyst stage was recorded. iBFC-cloned embryos were transferred by laparoscopy into the oviducts of DSH recipients after induced ovulation or oocyte aspiration (Gómez et al. 2004).
Detection of cat gene transcripts by RT-qPCR
The expression of Oct-4, Nanog and C-Myc and the internal standard GAPDH gene were detected in individual iBFC-cloned blastocysts by RT-qPCR as described previously (Gómez et al. 2010, 2011). The delta-delta Ct method was used for real-time PCR data evaluation. Data were collected using iQ5 Optical System Software (version 1.0; BioRad, Hercules, CA, USA). Abundance levels of each gene were expressed as n-fold increase relative to DSH-IVF blastocysts. Two to seven blastocysts were analysed per each treatment.
Quantification of global DNA methylation in iBFC-cloned embryos
The fluorescence intensity of 5-meC in iBFC-cloned and DSH-IVF embryos at 19 h after activation or IVF was assessed by fluorescence microscopy according to the method previously reported by Carambula et al. (2009). Embryos from all treatments were incubated with primary anti-5-MeC (1 : 1000) and secondary sheep anti-mouse-FITC conjugated (1 : 128) antibodies. Appropriate controls for autofluorescence and non-specific binding were generated. The average profile was generated as the ratio of fluorescence intensity of 5-MeC/DNA signals to determine differences among embryos and/treatments. Five to 12 embryos were examined in each treatment.
One-way or two-way anova were used to analyse the ratio of fluorescence intensity of H3K9ac and 5-MeC in cells and embryos, interactions between cell and embryo treatments, to analyse the percentages of apoptotic cells, and differences between the 2−∆∆Ct values for each gene at each embryo treatment, and the data on the number of embryos that cleaved and developed to blastocyst stage. The Holm-Sidak or Dunn's methods were used to determine differences between two means after anova. Statistics were performed by using SigmaStat (version 3.1.1; Systat Software Inc., Point Richmond, CA, USA). Differences were considered significant at p < 0.05.
To explore epigenetic changes in the chromatin of BFC fibroblasts, cells were untreated or treated with 50 nm TSA or 500 nm SAID for 20 h and the relative levels of H3K9ac assessed by FCM. Results showed that TSA and SAID increased the acetylation levels (p < 0.003; Fig. 1a), but levels were higher with 500 nm SAID than with 50 nm TSA (p < 0.008; Fig. 1a). The optimal concentration of 5-aza-dC and incubation time required for reducing the global DNA methylation levels in BFC fibroblasts was evaluated by treating the cells with 1 and 10 nm of 5-aza-dC for 20 or 72 h. Analysis revealed that global DNA methylation levels were significantly reduced when cells were treated for 20 h, regardless of 5-aza-dC concentration, as compared to that of untreated cells (p < 0.05; Fig. 1b). In contrast, global DNA methylation levels in cells treated for 72 h with 1 or 10 nm of 5-aza-dC were not modified and were comparable to that of untreated cells (Fig. 1b).
To determine whether treating BFC fibroblasts with TSA, SAID or 5-aza-dC provoked cell cytotoxicity, the numbers of apoptotic cells were evaluated by FCM. Results indicated that neither TSA (8.5%) nor SAID (9.4%) provoked cytotoxicity in comparison with untreated cells (5.6%, Fig. 1c). In contrast, higher doses of 5-aza-dC (10 nm) increased the numbers of apoptotic cells, regardless of the incubation times (20 h = 40.4%; 72 h = 48.4%) as compared to that of BFC cells treated with 1 nm of 5-aza-dC for 20 h (35.8%) and 72 h (27.0%; p < 0.05; Fig. 1d), respectively.
To explore changes in the global levels of DNA methylation in iBFC-cloned embryos, we measured the 5-MeC fluorescence intensity in individual cloned embryos that were reconstructed with donor cells: (i) untreated and (ii) treated with 500 nm SAID and 1 nm 5-aza-dC for 20 h, and embryos: (i) continuously treated with 500 nm SAID for 20 h, (ii) continuously treated with 500 nm SAID and 1 nm 5-aza-dC for 20 h and (iii) untreated. The 5-Mec fluorescence intensity in cloned embryos was affected by embryo treatment (p = 0.034), but was not influenced by cell treatment (p = 0.434), and there was not an interaction between cell and embryo treatments (p = 0.972). Lower levels of 5-MeC were observed in cloned embryos treated with a combination of SAID and 5-aza-dC (0.517 ± 0.06), regardless of cell treatment, and untreated cloned embryos (0.479 ± 0.06), than that of cloned embryos treated with SAID (0.736 ± 0.07; p = 0.017; Fig. 2).
To determine whether gene expression is influenced by epigenetic-modifying compounds, we measured expression of Oct-4, Nanog and C-Myc in individual iBFC-cloned blastocysts that were reconstructed with donor cells untreated or treated according to 'Experiment 1' and embryos treated as in 'Experiment 2'. The RT-qPCR analysis revealed that a higher number of cloned blastocysts reconstructed with treated donor cells expressed Oct-4 (n = 10/16; 62.5%), Nanog (n = 5/16; 31.2%) and C-Myc (n = 14/16; 87.5%), than cloned blastocysts reconstructed with untreated donor cells and continuous treatment with SAID (Oct-4 =2/6; 33.3%, Nanog =0/6; 0%, C-Myc =1/6; 16.6%) or untreated cloned blastocysts (n = 2); where none of the genes were expressed (p < 0.05; Fig. 3). Oct-4 and C-Myc were up-regulated in all cloned embryos treated, despite of donor cell treatment and the relative abundance of Oct-4 and C-Myc mRNA were significantly higher when compared with the expression of DSH-IVF embryos (p < 0.05; Fig. 3), with the exception of cloned blastocysts where both donor cells and embryos were treated with SAID and 5-dc-AZA, where the relative expression of Oct-4, Nanog and C-Myc were similar to that of DSH-IVF embryos (Fig. 3).
For evaluating in vitro developmental competence, BFC cells were untreated or treated according to 'Experiment 1' and embryos treated as in 'Experiment 2'. Data showed that development of cloned embryos to the blastocyst stage was enhanced by the cell treatment (p = 0.002), and there was an interaction between cell and embryo treatment (p = 0.037), but development to blastocyst stage was not influenced by embryo treatment (p = 0.166; Table 1). Higher percentages of blastocysts were observed when embryos were reconstructed with BFC cells treated compared with embryos reconstructed with untreated cells (8.164 ± 1.45 vs 2.357 ± 1.87, respectively; p < 0.05), and when cloned embryos were not continuously treated (12.525 ± 1.93; Table 1). We then tested in vivo developmental competence of iBFC-cloned embryos that were reconstructed with treated cells and embryos were (i) continuously treated with 500 nm SAID and 1 nm 5-aza-dC or (ii) untreated. Pregnancies in recipient cats were established with both types of embryos but foetuses started reabsorbing after 45 days of gestation (Table 2).
Table 1. Embryo cleavage (Day 2) and development to blastocyst stage (Day 8) of interspecies black-footed cat cloned embryos
Embryos cultured n
Cleavage n/total fused (mean% ± SEM)
Blastocysts n/total cleaved (mean% ± SEM)
Different superscripts within the same column indicate significant differences (p < 0.05).
Table 2. Number of pregnancies, embryo implantation and foetuses reabsorbed of interspecies black-footed cat cloned embryos
Pregnant n (%)
Per recipient ±SD
Implanted n (%)
Foetus reabsorbed n (%)
SAID + 5-aza-dC
SAID + 5-aza-dC
34.5 ± 12.2
67.5 ± 9.1
We demonstrated that epigenetic changes in BFC chromatin can be induced by treating the cells with a combination of 500 nm of SAID and 1 nm of 5-aza-dC for 20 h, which was the optimal combination to allow an increase in the acetylation levels in H3K9 and a reduction in the global DNA methylation levels without inducing cytotoxicity. Treatment of cells positively affected the efficiency of SCNT by enhancing up to fivefold the blastocyst rates, and by up-regulating the expression of Oct-4, Nanog and C-Myc genes at the blastocyst stage. Further improvements were observed when continuous treatment of cells and embryos with SAID and 5-aza-dC were applied, resulting in activation of Oct-4, Nanog and C-Myc to levels that resemble the gene expression pattern in DSH-IVF embryos. These results give additional credence to the concept that adequate epigenetic modifications to the chromatin of BFC cells and cloned embryos are important steps for enhancing reprogramming and that there is a synergistic effect between SAID and 5-aza-dC for the restoration of expression of pluripotent genes at the blastocyst stage.
The ‘correcting’ effect on gene expression has been also reported in bovine cloned embryos treated with TSA and 5-aza-dC, where the expression levels of Oct-4, Sox-2 and IGF2 were regulated to similar levels to that of their IVF counterpart embryos (Wang et al. 2011a). The authors suggested that TSA may act synergistically with 5-aza-dC by not only ‘correcting’ the gene expression but also by reducing the DNA methylation levels on satellite I region in blastocysts, resulting in enhancement of in vitro and in vivo developmental competence of treated cloned embryos (Wang et al. 2011a,b). Likewise, we observed that global DNA methylation was reduced when embryos were treated with SAID and 5aza-dC, regardless of cell treatment. However, in vitro developmental competence was not enhanced when both cells and embryos were treated, in contrast to the response when only donor cells were treated. It is possible that the low concentration of 5-aza-dC and/or length of treatment time of cloned embryos were not sufficient to induce further reduction in global DNA methylation levels that may be reflected indirectly in higher in vitro developmental competence. We choose to treat cloned embryos with a dose of 1 nm of 5-aza-dC because of the cytotoxic effect observed when donor cells were treated with 10 nm, and we maintained the cloned embryos in 5-aza-dC up to 20 h because our preliminary studies showed that treatment of cloned embryos with 1 nm or 5 nm of 5-aza-dC for 48 and 96 h induced cytotoxic effects as indicated by a reduced cleavage rate and a failure to develop to the blastocyst stage (data not shown). Further studies testing less toxic inhibitors of DNA methylation and a better understanding of how these inhibitors interact with DNA methyltransferases, will help to discover additional treatments for modifying epigenetic events in cloned embryos and enhance SCNT efficiency.
Reactivation of the pluripotent gene Nanog is an important step towards nuclear reprogramming. In felids, Nanog is a key intrinsic determinant for maintaining self-renewal of embryonic stem cells (ESC), and for derivation of induced pluripotent stem cells (Gómez et al. 2010; Verma et al. 2012). The mechanism(s) of how epigenetic-modifying compounds increase(s) and regulate(s) the expression of pluripotent genes in cloned embryos has not been elucidated. However, global loss of methylation in promoter regions in Oct-4 and Nanog genes in human cells, and H3K9 in mice PGCs was associated with reactivation of active genes (Deb-Rinker et al. 2005; Yamaguchi et al. 2005). It seems that changes in global DNA methylation per se is not the primary event in switching on gene transcription, but additional changes in methylation of H3K9 and of DNA cytosines and in promoter gene regions are conjunct events to give the chromatin an active state (Gidekel and Bergman 2002). Therefore, we suggest that reactivation of Oct-4 and Nanog in cloned blastocysts may be in part because of global losses of methylation in DNA and in H3K9 (Gómez et al. 2011) and possibly in the promoter regions of Nanog and Oct-4.
Regardless of the beneficial effects of treating cells and/or embryos with SAID and 5-aza-dC, in vivo developmental competence of cloned embryos was not enhanced. Similarly as previous studies, we observed the presence of amorphous mass of foetal tissue inside the gestational vesicles, and later death. However, foetal content inside the gestational sac in cloned embryos treated with both compounds was better organized had a larger mass size, and pregnancy was maintained for more than 45 days. We previously suggested that one possible reason for absorption of iBFC-cloned embryos may be due to the inability of the domestic cat to carry the pregnancy of BFC kittens. However, in a recent study (published in this Journal issue), we demonstrated that the domestic cat can carry to term the pregnancy of BFC after transfer of IVF derived embryos. Therefore, abnormal reprogramming of the BFC nucleus and possibly nucleocytoplasmic incompatibilities between the nucleus of the BFC and the cytoplasm of the domestic cat may be affecting the quality of derived embryos and resulting in pregnancy losses. Additional studies to produce cloned embryos using BFC oocytes as recipient cytoplasts may help to elucidate if incompatibilities between both species are affecting the viability of resultant iBFC-cloned embryos.
We are grateful to Dr. Bob McLean and Amanda Franklin for the surgical procedures and to animal care personnel at ACRES for the care of domestic cats. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the US Government.
Conflict of interest
None of the authors have any conflict of interest to declare.