Abstract
Nuclear transfer (NT) is a technique used to investigate the development and reprogramming potential of a single cell. DNA methyltransferase-3-like, which has been characterized as a repressive transcriptional regulator, is expressed in naturally fertilized egg and morula/blastocyst at pre-implantation stages. In this study, we demonstrate that the use of Dnmt3l-knockout (Dnmt3l-KO) donor cells in combination with Trichostatin A treatment improved the developmental efficiency and quality of the cloned embryos. Compared with the WT group, Dnmt3l-KO donor cell-derived cloned embryos exhibited increased cell numbers as well as restricted OCT4 expression in the inner cell mass (ICM) and silencing of transposable elements at the blastocyst stage. In addition, our results indicate that zygotic Dnmt3l is dispensable for cloned embryo development at pre-implantation stages. In Dnmt3l-KO mouse embryonic fibroblasts, we observed reduced nuclear localization of HDAC1, increased levels of the active histone mark H3K27ac and decreased accumulation of the repressive histone marks H3K27me3 and H3K9me3, suggesting that Dnmt3l-KO donor cells may offer a more permissive epigenetic state that is beneficial for NT reprogramming.
Introduction
Nuclear transfer (NT) is one of several approaches that have been developed to reprogram terminally differentiated somatic cells toward pluripotency or totipotency in an effort to better understand epigenetic reprogramming and generate in vitro models for disease research (Nishikawa et al. 2008, Jullien et al. 2011). NT consists of transferring a differentiated somatic cell into an enucleated oocyte to reacquire totipotency. Embryonic stem cells generated from NT-derived blastocysts (NT-ESCs) serve as a potential resource for regenerative medicine. However, the low efficiency and developmental abnormalities of cloned embryos remain important challenges in NT-associated research and applications (Yang et al. 2007, Rodriguez-Osorio et al. 2012, Long et al. 2014).
Insufficient reprogramming of the genome is thought to be a critical cause of the developmental failure of cloned embryos. Compared to fertilized embryos, cloned embryos display higher levels of DNA methylation and incomplete chromatin remodeling (Zhang et al. 2009, Mason et al. 2012, Rodriguez-Osorio et al. 2012). To improve NT efficiency, several approaches, including the use of chemical inhibitors such as HDAC inhibitors (HDACi) and 5-aza-C, have been devised to facilitate the reprogramming of somatic nuclei and establish embryonic epigenetic patterns (Ogura et al. 2013). HDACi treatment for altering the epigenetic status of donor cells or cloned embryos can significantly increase cloning efficiency. Treatment of donor cell with HDACi, trichostatin A (TSA), increases the developmental efficiency and quality of the resulting cloned embryos (Enright et al. 2003, Enright et al. 2005, Ding et al. 2008), suggesting that the initial chromatin status of the somatic cell contributes to the development of the cloned embryo. Furthermore, reduction of the repressive mark H3K9me3 in donor cells also significantly increases cloning efficiency, indicating that the global reduction of the repressive histone modification is beneficial to the onset of reprogramming in cloned embryos (Antony et al. 2013, Matoba et al. 2014). In addition to treatments of donor cells, TSA-treated cloned embryos also displayed lower repressive epigenetic marks, enhanced histone acetylation, remodeling of pericentromeric heterochromatin and increased nascent RNA transcription (Kishigami et al. 2006, Rybouchkin et al. 2006, Maalouf et al. 2009, Bui et al. 2010). These results suggest that establishing an accessible chromatin state can significantly improve genomic reprogramming.
DNA methyltransfer-3-like (DNMT3L) is an epigenetic factor that contributes to the establishment of DNA methylation and histone modifications (Chedin et al. 2002, Suetake et al. 2004, Kao et al. 2014). DNMT3L lacks catalytic activity but acts as a co-factor for the establishment of CG methylation and non-CG methylation (Ichiyanagi et al. 2013, Shirane et al. 2013, Vlachogiannis et al. 2015). DNMT3L can interact with histone H3 tail when its lysine 4 is unmethylated and induce de novo DNA methylation (Jia et al. 2007, Ooi et al. 2007). In addition, DNMT3L can recruit HDAC1 into the nucleus to perform histone deacetylation (Aapola et al. 2002, Deplus et al. 2002) and can facilitate HDAC1/TRIM28/SETDB1 complex formation in the nucleus, which accompanies the accumulation of repressive chromatin modification and transcriptional repression without triggering de novo DNA methylation (Kao et al. 2014).
Accumulating data indicate that DNMT3L is important for the dosage-dependent regulation of imprinted genes and transposable element silencing in germ cell development (Bourc'his & Bestor 2004, Kato et al. 2007, Lucifero et al. 2007, Smallwood et al. 2011, Kobayashi et al. 2012, Hara et al. 2014). During the pre-implantation period, Dnmt3l expression is detected in naturally fertilized eggs but subsequently declines until reaching the morula/blastocyst stage, at which it dramatically increases (Vassena et al. 2005). A recent study demonstrated that deletion of zygotic Dnmt3l resulted in delayed DNA methylation in naturally fertilized embryos (Guenatri et al. 2013). Furthermore, our previous results revealed that DNMT3L facilitates interactions of the transcriptional repressors DNMT3A, SETDB1, TRIM28 and HDAC1 in embryonic stem cells (Kao et al. 2014). Zygotic Dnmt3l expression commences at the morula/blastocyst stage, similar to the H3K9me3 methyltransferase SETDB1 (Dodge et al. 2004, Vassena et al. 2005). SETDB1 is recognized as a transcriptional repressor that can interact with HDAC1 and TRIM28 to silence gene expression (Schultz et al. 2002, Yang et al. 2003). Hence, removal of zygotic Dnmt3l/DNMT3L may disrupt epigenetically mediated gene silencing and result in more accessible chromatin for nuclear reprogramming, and therefore improve the developmental efficiency of NT embryos.
In this study, we used mouse embryonic fibroblasts (MEFs)-derived from Dnmt3l-knockout (Dnmt3l-KO) mice as donor cells for NT to examine the role of DNMT3L in cloned embryo development. Our results revealed a synergistic effect between Dnmt3l-KO and treatment with TSA to enhance the developmental efficiency and quality of cloned embryos at the blastocyst stage. Our data revealed that zygotic Dnmt3l expression was dispensable for cloned embryo development during pre-implantation stages. Instead, we observed that Dnmt3l-KO donor cells displayed decreased levels of H3K9me3 and H3K27me3, higher levels of the active histone mark H3K27ac and increased cytoplasmic localization of HDAC1, indicative of a permissive epigenetic state beneficial for nuclear reprogramming.
Materials and methods
Animals, oocyte collection and donor cell preparation
Animal care and protocols used throughout the study were approved by the Institutional Animal Care and Use Committee (IACUC) of National Taiwan University. C57BL/6 (B6) female mice were bred with DBA/2 males to obtain the hybrid strain B6D2F1 (C57BL/6J×DBA/2J). For NT experiment, the oocytes at the MII stage used were harvested from female B6D2F1 mice. Superovulation of 8- to 12-week-old mice was induced with 7.5 IU of pregnant mare serum gonadotropin followed by 7.5 IU of human chorionic gonadotropin (hCG) 48 h later. The oocytes were harvested 13–14 h after hCG treatment. The cumulus–oocyte complexes were incubated with 0.1 mg/ml hyaluronidase in M2 medium for 3 min and then washed in CZB-HEPES medium by gentle pipetting with glass pipette to remove the cumulus cells.
For donor cell preparation, C57BL/6 (B6) WT and Dnmt3l-KO MEF cell lines were derived from day 13.5–14.5 embryos via Dnmt3l heterozygote (Dnmt3l+/−) intercrosses (Hata et al. 2002). MEFs from passages four to six were used for NT. Near-confluent MEFs derived from WT and Dnmt3l-KO embryos were cultured under serum starvation conditions (0.1% FBS in DMEM) for 5 days. The starved MEFs were trypsinized, incubated with 10% FBS in DMEM and centrifuged to obtain single cell for NT.
NT and embryo culture
NT was performed as previously described (Sung et al. 2010). Briefly, cumulus-free oocytes were held with a holding micropipette in CZB-HEPES medium containing 5 μg/ml cytochalasin B (CB), and the spindle chromosome complexes were removed with a pipette with an inner diameter of 10 μm using a Piezo-drill micromanipulator (PMAS-CT 150, Prime-Tech, Tsuchiura, Japan). The nuclear donor of MEFs with a diameter of 18–20 μm was selected and we inserted the donor cell into the perivitelline space of an enucleated mouse oocyte. After transfer, the cell–cytoplast complexes were induced to fuse with two DC pulses of 150 V/1 mm for 10 μs using an Electro Cell Manipulator 200 (BTX, San Diego, CA, USA). The reconstituted oocytes were activated in calcium-free CZB medium containing 10 mM SrCl, 5 μg/ml cytochalasin B and 10 nM TSA for 6 h, followed by culture with 10 nM TSA in KSOM+AA medium (MR-121-D, Millipore, Billerica, MA, USA) for 4 h. The reconstructed embryos were cultured in KSOM+AA medium at 37 °C in 5% CO2 humidified air.
Immunofluorescence staining
The embryos were fixed with 4% paraformaldehyde (PFA), washed with 0.1% polyvinyl alcohol in PBS (PVA-PBS) and treated with 0.5% Triton X-100 for 30 min. After incubation with 0.25% Tween-20 in PBS for 30 min, the embryos were blocked with 3% BSA in PVA-PBS for 2 h and then washed with 0.1% PVA-PBS at room temperature. The embryos were incubated at 4 °C overnight with the following primary antibodies: anti-OCT4 (POU5F1) (sc-9081; Santa Cruz Biotechnology, 1:1000 dilution) and anti-CDX2 (MU392A-UC; BioGenex, San Ramon, CA, USA, 1:1000 dilution). After incubation with donkey anti-rabbit-594 (A21207, Invitrogen, 1:500 dilution) and goat anti-mouse 488 (A11029, Invitrogen, 1:500 dilution) secondary antibodies, the embryos were counterstained with DAPI and mounted with a mounting medium.
For immunocytochemistry analysis, the cultured MEFs were fixed in 4% PFA, washed with 0.1% Tween-20 in PBS (PBST) and treated with 0.25% Triton X-100 in PBS. After washing with PBST, the cells were blocked with 2% BSA in PBST for 1 h at room temperature and incubated at 4 °C overnight with an anti-HDAC1 primary antibody (ab7028; Abcam, Cambridge, MA, USA, 1:5000 dilution). The cells were then washed in PBST, incubated with secondary antibodies, stained with Hoechst33342 (Sigma), mounted with a mounting medium (P36934, Invitrogen) and analyzed using a Leica TCS SP5 II confocal microscope.
cDNA amplification and quantitative PCR
The approach utilized for global cDNA amplification and quantitative PCR (qPCR) from a single embryo or oocyte followed that of Kurimoto et al. (2007). In brief, the single embryo at different stages or the oocyte was lysed without purification. First-strand cDNAs were synthesized using a poly(dT)-tailed primer followed by exonuclease treatment to specifically eliminate unreacted primers. The second strands were generated with a second poly(dT)-tailed primer after poly(dA) tailing of the first-strand cDNAs. The cDNAs were amplified by PCR and the products were used for transcript expression analyses by qPCR. The monocolor hydrolysis probe detection system (Roche Diagnostics) was used to quantitate mRNA expression levels. The qPCR primers and corresponding probe numbers from Universal Probe Library (UPL) are as follows: Rplp0-forward, cttcccacttactgaaaaggtca, and Rplp0-reverse, tcctttgcttcagctttgg (UPL probe no. 72); Dnmt3l-forward, agtcagaagcaggagcaagc, and Dnmt3l-reverse, agcgggagaaggcagttc (UPL probe no. 79); MuERVL-forward, ggtatgggtcaatcctccag, and MuERVL-reverse, tcctctactacccattctgtatttcc (UPL probe no. 41); and IAP-forward, tgaaggttcagtgtcctagttcc, and IAP-reverse, acaggcttttacccagagca (UPL probe no. 32). These primer sequences are also listed in Table S1 of the supplementary material, see section on supplementary data given at the end of this article. qPCR was performed in a thermal cycler (LightCycler 480 II Instrument; Roche Applied Science) using the following program: 50 °C for 2 min, 95 °C for 10 min, and 40 cycles of denaturation at 95 °C for 15 s and annealing and extension at 60 °C for 1 min. The housekeeping gene Rplp0 was used as a normalization control. The 2−ΔCp method was used to quantify the qPCR results.
DNA extraction and bisulfite sequencing
Genomic DNA was extracted from MEFs at passage 5 (P5) with a DNeasy Blood and Tissue kit (#69504, Qiagen). The genomic DNA was subjected to bisulfite treatment using an EZ DNA methylation kit according to the manufacturer's instructions (Zymo Research, Orange, CA, USA), followed by PCR amplification and PCR product purification using a QIAquick Gel Extraction Kit (#28704, Qiagen). The PCR products were cloned using the pGEM-T Easy kit for subsequent sequencing (Promega). The sequencing results were analyzed using a BiQ Analyzer (Bock et al. 2005).
Quantitative RT-PCR
WT and Dnmt3l-KO MEFs were collected and treated with TRIzol (Invitrogen) to extract RNA. Reverse transcription reactions were performed using the SuperScript First-Strand Synthesis System (Invitrogen). qPCR was performed in a Roche Light Cycler 480II. The expression levels were detected using SYBR Green PCR master mix (KAPA Biosystems, Woburn, MA, USA) by the following program: 50 °C for 2 min, 95 °C for 10 min, and then 45 cycles of denaturation at 95 °C for 15 s and annealing and extension at 60 °C for 45 s. The 2−ΔΔCp method was used to quantify the qPCR results. Gapdh was used as a housekeeping gene for normalization. The primers used are listed in Table S1 of the supplementary material, see section on supplementary data given at the end of this article.
Western blotting analysis
Western blotting was performed according to a standard protocol (Sambrook & Russell 2001) using PVDF membranes (Millipore). The cultured MEFs were lysed in RIPA buffer (Millipore) supplemented with 1× protease inhibitor cocktail (591134, Millipore), 1× phosphatase inhibitor cocktail (524629, Millipore), and 1 mM PMSF. The membranes were blocked with a blocker (Bløk-PO or Bløk-CH, Millipore) and 5% BSA (Sigma). Incubations with the primary antibodies anti-HDAC1 (ab7028, Abcam, 1:10 000 dilution), anti-H3K9me3 (05-1242, Millipore, 1:2000 dilution), anti-H3K9me3 (ab8898, Abcam, 1:3000 dilution), anti-H3K9ac (07-352, Millipore, 1:2000 dilution), anti-H3K27me3 (05-1951, Millipore, 1:2000 dilution), anti-H3K27ac (ab4729, Abcam, 1:2000 dilution), anti-H3ac (06-599, Millipore, 1:2000 dilution), anti-H3K18ac (#9675, Cell Signaling Technology, Beverly, MA, USA, 1:2000 dilution), and anti-H3 (06-755, Millipore, 1:2000 dilution) were performed at 4 °C overnight with gentle shaking. After incubating with a horseradish peroxidase-conjugated secondary antibody (Pierce, Rockford, IL, USA, 1:10 000 dilution), the proteins were detected using a chemiluminescent detection reagent (Millipore). The images were quantified with ImageJ Software (http://rsbweb.nih.gov/ij/). The antibodies are listed in Table S2 of the supplementary material, see section on supplementary data given at the end of this article.
RNA sequencing and bioinformatics analysis
Total RNA for library preparation, sequencing, and RNA-seq analysis was extracted using TRIzol Reagent (Ambion, Austin, TX, USA). Following the manual, after extraction procedures, the total RNA was washed by 80% ethanol, and re-suspended in nuclease-free H2O. The concentration and quality of the RNA were determined using a RNA BR Chip (Agilent Technologies, Palo Alto, CA, USA) with an Agilent 2100 Bioanalyzer (Agilent Technologies). The isolated RNAs were rRNA depleted and purified by a RiboMinus Kit (Invitrogen). Following purification, the rRNA-free RNAs were fragmented into small pieces. The cleaved RNA fragments were copied into first-strand cDNA using random primers, followed by second-strand cDNA synthesis, a single A base addition and adapter ligation using a TruSeq RNA library preparation kit (Illumina, San Diego, CA, USA). The products were then enriched through 12 cycles of PCR amplification to create the final cDNA library. The concentration of final cDNA libraries were determined by a DNA HS Chip (Agilent Technologies) with an Agilent 2100 Bioanalyzer. The high throughput sequencing process was performed by Illumina HiSeq 2500 sequencer (Illumina, Inc., CIC bioGUNE). Notably, all of the cDNA libraries were pre-sequenced with typical TA-cloning methods using an Escherichia coli strand (DH5-α) to estimate the rRNA depletion rate before high throughput sequencing.
The adaptor-trimmed reads were aligned against the mm10 mouse reference genome using TopHat2 (v2.0.12) with sequence-specific parameter and were used for transcript fragment assembly with Cufflinks (v2.2.1) (Trapnell et al. 2012, Kim et al. 2013). After the assembly, transcript expression levels were calculated based on eXpress (v1.4.1) (Roberts & Pachter 2013) against the mm10 refseq database to identify differentially expressed genes and transcripts. For cross-sample comparison, expression levels were normalized and analyzed using DESeq2 (v1.8.0) (Love et al. 2014). Gene ontology (GO) analysis was performed using Panther Software (http://pantherdb.org/) (Mi et al. 2013). The sequences have been deposited in the NCBI GEO dataset under accession number GSE69007.
Results
The use of Dnmt3l-KO donor cells synergizes with TSA treatment to improve the developmental efficiency of the cloned embryos
To evaluate whether DNMT3L contributes to the developmental efficiency of reconstructed embryos, we analyzed cloned embryos generated with fibroblast donors from WT and Dnmt3l-KO embryos until 96 h post-activation. In the absence of the HDACi TSA treatment, most of the cloned embryos underwent two-cell division but then arrested at this stage (Fig. 1A and B; Table 1). Low percentages of cloned morula and blastocyst embryos were generated from WT MEFs (5±1.6% of morula stage and 3±2.3% of blastocyst stage) or Dnmt3l-KO MEF donor cells (15±6.1% of morula stage and 9±4.8% of blastocyst stage).
The developmental efficiency of cloned embryos derived from WT and Dnmt3l-KO MEF donor cells. Cloned embryos were collected 96 h post-activation. (A and B) Without TSA treatment, most cloned embryos were arrested at the two-cell stage, whether they were derived from WT or Dnmt3l-KO MEF donor cells. (C and D) Under TSA treatment, both WT and Dnmt3l-KO MEF-derived embryos displayed increased developmental efficiency compared with non-TSA treatment groups. In addition, a significant proportion of Dnmt3l-KO NT embryos developed to the blastocyst stage. Scale bar, 100 μm. (E) Quantification results indicated that the percentage of blastocysts derived from Dnmt3l-KO MEFs with TSA treatment was the highest compared with the other groups. The results are presented as the means±s.e.m. from at least three independent experiments. *denotes P<0.05, Student's t-test. w/o, without TSA treatment; w/TSA with TSA treatment. D3L-KO, Dnmt3l-KO.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
The developmental efficiency of cloned embryos derived from WT and Dnmt3l-KO MEF donor cells. Cloned embryos were collected 96 h post-activation. (A and B) Without TSA treatment, most cloned embryos were arrested at the two-cell stage, whether they were derived from WT or Dnmt3l-KO MEF donor cells. (C and D) Under TSA treatment, both WT and Dnmt3l-KO MEF-derived embryos displayed increased developmental efficiency compared with non-TSA treatment groups. In addition, a significant proportion of Dnmt3l-KO NT embryos developed to the blastocyst stage. Scale bar, 100 μm. (E) Quantification results indicated that the percentage of blastocysts derived from Dnmt3l-KO MEFs with TSA treatment was the highest compared with the other groups. The results are presented as the means±s.e.m. from at least three independent experiments. *denotes P<0.05, Student's t-test. w/o, without TSA treatment; w/TSA with TSA treatment. D3L-KO, Dnmt3l-KO.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
The developmental efficiency of cloned embryos derived from WT and Dnmt3l-KO MEF donor cells. Cloned embryos were collected 96 h post-activation. (A and B) Without TSA treatment, most cloned embryos were arrested at the two-cell stage, whether they were derived from WT or Dnmt3l-KO MEF donor cells. (C and D) Under TSA treatment, both WT and Dnmt3l-KO MEF-derived embryos displayed increased developmental efficiency compared with non-TSA treatment groups. In addition, a significant proportion of Dnmt3l-KO NT embryos developed to the blastocyst stage. Scale bar, 100 μm. (E) Quantification results indicated that the percentage of blastocysts derived from Dnmt3l-KO MEFs with TSA treatment was the highest compared with the other groups. The results are presented as the means±s.e.m. from at least three independent experiments. *denotes P<0.05, Student's t-test. w/o, without TSA treatment; w/TSA with TSA treatment. D3L-KO, Dnmt3l-KO.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
Developmental rate of cloned embryos.
| Donor cell type | TSA treatment | Number of activated oocytesa | Number of two-cell embryos (%) | Number of morula embryos (%) | Number of blastocyst embryos (%) |
|---|---|---|---|---|---|
| WT | − | 268 | 241 (90±3.0) | 13 (5±1.6)* | 9 (3±2.3) |
| + | 206 | 196 (95±1.5) | 94 (46±7.9)† | 51 (25±2.4)§ | |
| Dnmt3l-KO | − | 317 | 292 (92±4.0) | 48 (16±6.1)* | 30 (9±4.8)§ |
| + | 146 | 138 (95±4.6) | 115 (79±5.1)‡ | 99 (68±9.3)∥ |
M, morula embryos were check at day 4 after NT; BL, blastocyst embryos were check at day 5 after NT; the developmental rate was based on activated oocytes. The results are presented as the means±s.e.m. Statistical comparisons were determined by ANOVA with Duncan's multiple range test. Values with different superscripts are significantly different (P<0.05). * vs †, † vs ‡, § vs ∥.
These results were compiled from three experimental replicates.
To further investigate the potential relationship between Dnmt3l-KO donor cell and HDACi, we performed NT experiments in the presence of TSA. Notably, under TSA treatment, cloned embryos derived from Dnmt3l-KO donor cells exhibited significantly increased blastocyst formation efficiency (68%), which was 2.5-fold higher than the WT group (25%) (Fig. 1C, D and E; Table 1), indicating an improved NT efficiency in the Dnmt3l-KO cloned embryos under TSA treatment. Furthermore, these results suggest that Dnmt3l depletion and TSA treatment synergistically improve the efficiency of cloned embryos. In the following experiments, we focused on comparisons between the cloned embryos generated from WT and Dnmt3l-KO donor cells under TSA treatment.
Dnmt3l-KO donor cells improve the quality of cloned embryos under TSA treatment
To compare the quality of the cloned embryos derived from these two types of donor cells under TSA treatment, we calculated the total cell numbers of the cloned embryos at the blastocyst stage. Fertilized embryos were used as a positive control. The results indicated that the blastocysts derived from the Dnmt3l-KO donor cells exhibited significantly higher total cell numbers compared with the embryos generated from WT cells under TSA treatment (Fig. 2A). Next, we classified the cloned blastocysts from grade I to grade III based on their quality, represented by total cell numbers and OCT4/CDX2 expression patterns in the blastocysts (Fig. 2B). Grade I was defined as a blastocyst with more than 60 cells, with OCT4 expression restricted to the ICM and CDX2 expression limited to the trophectoderm. Grade II consisted of blastocysts with cell numbers ranging between 30 and 60; OCT4 was expressed in both the ICM and trophectoderm, and CDX2 was expressed in the trophectoderm. Grade III was characterized as a blastocyst with a low cell number (<30) and weak OCT4 and CDX2 expression patterns, as well as disorganized localization of the OCT4 signal and a lack of a clear ICM structure.
The quality of WT and Dnmt3l-KO cloned embryos under TSA treatment. (A) The use of Dnmt3l-KO MEFs increased the total cell number of cloned embryos at the blastocyst stage under TSA treatment. The bars indicate the means±s.e.m. from at least three independent experiments. (B) The blastocysts were classified into three classes, grades I–III, based on the expression and localization of OCT4 and CDX2 as well as total cell number. Scale bar=50 μm. (C) The percentage of different graded blastocysts derived from WT and Dnmt3l-KO MEFs. The results were calculated from at least three independent experiments. Parthenogenetic embryos were used as a control. N, number of embryos; D3L-KO, Dnmt3l-KO.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
The quality of WT and Dnmt3l-KO cloned embryos under TSA treatment. (A) The use of Dnmt3l-KO MEFs increased the total cell number of cloned embryos at the blastocyst stage under TSA treatment. The bars indicate the means±s.e.m. from at least three independent experiments. (B) The blastocysts were classified into three classes, grades I–III, based on the expression and localization of OCT4 and CDX2 as well as total cell number. Scale bar=50 μm. (C) The percentage of different graded blastocysts derived from WT and Dnmt3l-KO MEFs. The results were calculated from at least three independent experiments. Parthenogenetic embryos were used as a control. N, number of embryos; D3L-KO, Dnmt3l-KO.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
The quality of WT and Dnmt3l-KO cloned embryos under TSA treatment. (A) The use of Dnmt3l-KO MEFs increased the total cell number of cloned embryos at the blastocyst stage under TSA treatment. The bars indicate the means±s.e.m. from at least three independent experiments. (B) The blastocysts were classified into three classes, grades I–III, based on the expression and localization of OCT4 and CDX2 as well as total cell number. Scale bar=50 μm. (C) The percentage of different graded blastocysts derived from WT and Dnmt3l-KO MEFs. The results were calculated from at least three independent experiments. Parthenogenetic embryos were used as a control. N, number of embryos; D3L-KO, Dnmt3l-KO.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
The percentage of good quality (grade I) embryos was higher among the blastocysts derived from Dnmt3l-KO donor cells than in those derived from WT donor cells (Fig. 2C). There was no obvious difference in the percentage of grade II embryos between these two groups. In contrast, compared with WT, the percentage of blastocysts with poor development (grade III) was lower among the Dnmt3l-KO cloned embryos (Fig. 2C). These data suggest that the use of Dnmt3l-KO donor cells improved the quality of the cloned embryos.
Zygotic Dnmt3l is not required for development and transposable element regulation in cloned embryos during pre-implantation stages
To investigate whether zygotic Dnmt3l is involved in NT embryo development at pre-implantation stages, we examined the amount of Dnmt3l expression in single enucleated oocyte or cloned embryo by cDNA amplification and qRT-PCR analysis. Each nuclei-removed oocyte still contained high levels of Dnmt3l transcripts (Fig. 3A). After injecting the donor cell into the enucleated oocyte, we continued to detect Dnmt3l expression throughout pre-implantation development in the cloned embryos. At the two-cell stage, Dnmt3l transcripts were still readily detectable in most embryos derived from both WT and Dnmt3l-KO MEFs (Fig. 3A). However, Dnmt3l expression became quite low in both WT and Dnmt3l-KO cloned embryos after the two-cell stage. The expression level of Dnmt3l did not increase significantly, even in the WT cloned embryos, until the late blastocyst stage, indicating a lack of significant zygotic Dnmt3l expression in pre-implantation cloned embryos (Fig. 3A).
Expression of Dnmt3l and transposable elements in cloned embryos at pre-implantation stages. (A) Dnmt3l expression in cloned embryos. Single enucleated oocyte or embryo was analyzed by qPCR. Dnmt3l expression was readily detectable in both WT and Dnmt3l-KO MEF-derived cloned nuclei-removed oocytes but not in morulas and blastocysts. (B) MuERVL and IAP expression levels in individual cloned embryos by qPCR. No obvious difference in the MuERVL or IAP expression pattern was observed between the two types of cloned embryos at pre-implantation stages. Rplp0 was used as an internal control for mRNA expression analysis. Quantification of mRNA expression was performed using the 2−ΔCp method. Error bars represent the s.e.m. generated from independent embryos with three technical repeats. D3L-KO, Dnmt3l-KO.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
Expression of Dnmt3l and transposable elements in cloned embryos at pre-implantation stages. (A) Dnmt3l expression in cloned embryos. Single enucleated oocyte or embryo was analyzed by qPCR. Dnmt3l expression was readily detectable in both WT and Dnmt3l-KO MEF-derived cloned nuclei-removed oocytes but not in morulas and blastocysts. (B) MuERVL and IAP expression levels in individual cloned embryos by qPCR. No obvious difference in the MuERVL or IAP expression pattern was observed between the two types of cloned embryos at pre-implantation stages. Rplp0 was used as an internal control for mRNA expression analysis. Quantification of mRNA expression was performed using the 2−ΔCp method. Error bars represent the s.e.m. generated from independent embryos with three technical repeats. D3L-KO, Dnmt3l-KO.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
Expression of Dnmt3l and transposable elements in cloned embryos at pre-implantation stages. (A) Dnmt3l expression in cloned embryos. Single enucleated oocyte or embryo was analyzed by qPCR. Dnmt3l expression was readily detectable in both WT and Dnmt3l-KO MEF-derived cloned nuclei-removed oocytes but not in morulas and blastocysts. (B) MuERVL and IAP expression levels in individual cloned embryos by qPCR. No obvious difference in the MuERVL or IAP expression pattern was observed between the two types of cloned embryos at pre-implantation stages. Rplp0 was used as an internal control for mRNA expression analysis. Quantification of mRNA expression was performed using the 2−ΔCp method. Error bars represent the s.e.m. generated from independent embryos with three technical repeats. D3L-KO, Dnmt3l-KO.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
These results suggest that the upregulation of zygotic Dnmt3l is not required for cloned embryo development at pre-implantation stages and that the improved quality of Dnmt3l-KO MEF-derived cloned embryos may not be primarily attributed to differences in Dnmt3l expression.
Given that proper regulation of transposable elements (TEs) is correlated with pre-implantation embryo development (Kigami et al. 2003, Peaston et al. 2004, Kim et al. 2014) and overexpression of the long terminal repeat retrotransposons, murine endogenous retroviruses-like (MuERVL) and intracisternal A-particle (IAP) retrotransposon are detrimental in stem cell pluripotency, we further investigated the expression levels of MuERVL and IAP in cloned embryo. We detected no obvious differences in the expression of MuERVL or IAP between WT and Dnmt3l-KO cloned embryos from the two-cell stage to the blastocyst stage (Fig. 3B). Therefore, during pre-implantation stages, although no obvious zygotic Dnmt3l expression was detected, MuERVL and IAP expressions were still repressed at the morula and blastocyst stages in most embryos, suggesting that zygotic Dnmt3l may not be required for TE repression in cloned embryos (Fig. 3).
Because zygotic Dnmt3l does not appear to be significantly involved in cloned embryo development at pre-implantation stages, we speculated that different properties of the WT and Dnmt3l-KO donor cells may contribute to the differential reprogramming efficiencies and qualities of the respective cloned embryos. Therefore, we further characterized potential facilitative and obstructive factors in the donor cells.
Reduced HDAC1 nuclear localization in Dnmt3l-KO MEF cells
Accumulating results indicate suppression of chromatin-modifying activities of histone deacetylases, including HDAC1, contribute to NT efficiency improvement (Enright et al. 2003, Rybouchkin et al. 2006, Ma & Schultz 2008). We thus investigated the subcellular localization of HDAC1 in donor cells by immunocytochemistry. Two types of HDAC1 staining were observed in the MEFs. In the cell with Type I pattern, HDAC1 expression was restricted to the nucleus, whereas the Type II cell exhibited both nuclear and cytoplasmic HDAC1 localization (Fig. 4A). Our assays demonstrated that HDAC1 was largely restricted to the nucleus in WT MEFs (Type I). In contrast, a significantly increased percentage of Dnmt3l-KO MEFs exhibited both nuclear and cytoplasmic localization of HDAC1 (Type II) compared with WT MEFs (Fig. 4B and C). We further examined the protein levels of HDAC1 via western blotting, which revealed no difference in the total amount of HDAC1 protein between WT and Dnmt3l-KO MEFs (Fig. 4D), indicating that the reduced HDAC1 staining signal in Dnmt3l-KO MEFs was primarily due to translocation to the cytoplasm.
HDAC1 subcellular localization and the relative levels of histone marks in WT and Dnmt3l-KO mouse embryonic fibroblast cells. (A) Two patterns of HDAC1 localization were observed in the MEFs: Type I (nuclear localization) and Type II (nuclear and cytoplasmic localization). Scale bar, 10 μm. (B) Immunostaining of HDAC1 in WT and Dnmt3l-KO MEFs. Scale bar, 50 μm. (C) A significantly higher percentage of Type II HDAC1 expression was observed in Dnmt3l-KO MEFs compared with WT MEFs. The error bars represent the s.e.m. generated from three independent experiments. (D) No obvious change in total HDAC1 protein expression was observed between WT and Dnmt3l-KO MEFs. GAPDH was used as an internal control for protein expression. Mean±s.e.m. were calculated from at least three biological replicates. (E) Western blotting of histone marks in WT and Dnmt3l-KO MEFs. Increased H3K27ac and reduced H3K9me3 and H3K27me3 levels were observed in Dnmt3l-KO MEFs. Histone H3 was used as an internal control. (F) Bar chart displaying the relative levels of active and repressive histone marks in WT and Dnmt3l-KO MEFs. The results are the mean±s.e.m. from at least three biological replicates. D3L-KO, Dnmt3l-KO. *denotes P<0.05 (Student's t-test).
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
HDAC1 subcellular localization and the relative levels of histone marks in WT and Dnmt3l-KO mouse embryonic fibroblast cells. (A) Two patterns of HDAC1 localization were observed in the MEFs: Type I (nuclear localization) and Type II (nuclear and cytoplasmic localization). Scale bar, 10 μm. (B) Immunostaining of HDAC1 in WT and Dnmt3l-KO MEFs. Scale bar, 50 μm. (C) A significantly higher percentage of Type II HDAC1 expression was observed in Dnmt3l-KO MEFs compared with WT MEFs. The error bars represent the s.e.m. generated from three independent experiments. (D) No obvious change in total HDAC1 protein expression was observed between WT and Dnmt3l-KO MEFs. GAPDH was used as an internal control for protein expression. Mean±s.e.m. were calculated from at least three biological replicates. (E) Western blotting of histone marks in WT and Dnmt3l-KO MEFs. Increased H3K27ac and reduced H3K9me3 and H3K27me3 levels were observed in Dnmt3l-KO MEFs. Histone H3 was used as an internal control. (F) Bar chart displaying the relative levels of active and repressive histone marks in WT and Dnmt3l-KO MEFs. The results are the mean±s.e.m. from at least three biological replicates. D3L-KO, Dnmt3l-KO. *denotes P<0.05 (Student's t-test).
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
HDAC1 subcellular localization and the relative levels of histone marks in WT and Dnmt3l-KO mouse embryonic fibroblast cells. (A) Two patterns of HDAC1 localization were observed in the MEFs: Type I (nuclear localization) and Type II (nuclear and cytoplasmic localization). Scale bar, 10 μm. (B) Immunostaining of HDAC1 in WT and Dnmt3l-KO MEFs. Scale bar, 50 μm. (C) A significantly higher percentage of Type II HDAC1 expression was observed in Dnmt3l-KO MEFs compared with WT MEFs. The error bars represent the s.e.m. generated from three independent experiments. (D) No obvious change in total HDAC1 protein expression was observed between WT and Dnmt3l-KO MEFs. GAPDH was used as an internal control for protein expression. Mean±s.e.m. were calculated from at least three biological replicates. (E) Western blotting of histone marks in WT and Dnmt3l-KO MEFs. Increased H3K27ac and reduced H3K9me3 and H3K27me3 levels were observed in Dnmt3l-KO MEFs. Histone H3 was used as an internal control. (F) Bar chart displaying the relative levels of active and repressive histone marks in WT and Dnmt3l-KO MEFs. The results are the mean±s.e.m. from at least three biological replicates. D3L-KO, Dnmt3l-KO. *denotes P<0.05 (Student's t-test).
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
Increased H3K27ac and decreased H3K27me3 and H3K9me3 levels in Dnmt3l-KO MEF cells
Accumulating evidence suggests that the epigenetic state of the donor cell correlates with the developmental efficiency of cloned embryos. Enhanced acetylation modifications on the histone tail improve cloning efficiency, whereas H3K9me3 is recognized as a critical epigenetic barrier in nuclear reprogramming (Enright et al. 2003, Antony et al. 2013, Matoba et al. 2014). We thus examined the levels of several histone marks, including the active marks H3K4me3, H3ac, H3K9ac, H3K18ac and H3K27ac and the repressive marks H3K9me3 and H3K27me3, implicative of the global chromatin state in the two types of donor cells. We did not observe obvious differences in the levels of H3K4me3, H3ac, H3K9ac, H3K18ac between WT and Dnmt3l-KO MEFs (Fig. 4E and F). However, compared with WT MEFs, Dnmt3l-KO MEFs exhibited significantly increased H3K27ac and decreased H3K9me3 and H3K27m3 levels (Fig. 4E and F). These data suggest that Dnmt3l-KO MEFs may present a more accessible chromatin state that facilitates NT reprogramming.
Increased expression of development-related genes in Dnmt3l-KO MEF cells
From our RNA-seq gene expression profiles of WT and Dnmt3l-KO MEFs, we found that 54 genes were differentially expressed between WT and Dnmt3l-KO MEFs (fold change >1.5 and P value <0.05) (Fig. 5A), of which 41 were obviously upregulated in Dnmt3l-KO MEFs. The differentially expressed genes between WT and Dnmt3l-KO MEFs were enriched for the Gene Ontology molecular function terms of nucleic acid binding, catalytic activity and nucleic acid binding transcription factor activity (Fig. 5B). In addition, we found that the expression levels of several genes associated with embryonic development, including Adam19, Chd8, Git1, Lims1/Pinch1, Pogz and Sulf1, were increased in Dnmt3l-KO MEFs (Nishiyama et al. 2004, Zhou et al. 2004, Liang et al. 2005, Holst et al. 2007, Pang et al. 2009). We further investigated the DNA methylation state of CpG islands in the promoter regions of three differentially expressed genes Adam19, Git1 and Lims1. The bisulfite sequencing results revealed that hypomethylation of the CpG islands in the promoter regions of these genes in both WT and Dnmt3l-KO MEFs (Supplementary Figure S1, see section on supplementary data given at the end of this article), suggesting that DNA methylation may not be the primary regulation of these genes in MEFs.
Increased expression of development-associated genes in Dnmt3l-KO MEFs. (A) Comparison of the expression profiles of WT and Dnmt3l-KO MEFs. Two biological replicates of WT and Dnmt3l-KO samples were analyzed. The red and green circles indicate genes that were differentially expressed between WT and Dnmt3l-KO MEFs (fold change >1.5 and P value <0.05). FC, fold change; pVal, P value; KO, Dnmt3l-KO. (B) Gene ontology (GO) term analysis of the differentially expressed genes in WT and Dnmt3l-KO MEFs. (C) Bar chart displaying the relative expression levels of differentially expressed genes in WT and Dnmt3l-KO MEFs. The qPCR results were quantified by 2−ΔΔCp method and Gapdh was used as a housekeeping gene for normalization. The error bars represent the s.e.m. from four to five biological repeats. *denotes P<0.05 and **indicates P<0.01 (Student's t test).
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
Increased expression of development-associated genes in Dnmt3l-KO MEFs. (A) Comparison of the expression profiles of WT and Dnmt3l-KO MEFs. Two biological replicates of WT and Dnmt3l-KO samples were analyzed. The red and green circles indicate genes that were differentially expressed between WT and Dnmt3l-KO MEFs (fold change >1.5 and P value <0.05). FC, fold change; pVal, P value; KO, Dnmt3l-KO. (B) Gene ontology (GO) term analysis of the differentially expressed genes in WT and Dnmt3l-KO MEFs. (C) Bar chart displaying the relative expression levels of differentially expressed genes in WT and Dnmt3l-KO MEFs. The qPCR results were quantified by 2−ΔΔCp method and Gapdh was used as a housekeeping gene for normalization. The error bars represent the s.e.m. from four to five biological repeats. *denotes P<0.05 and **indicates P<0.01 (Student's t test).
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
Increased expression of development-associated genes in Dnmt3l-KO MEFs. (A) Comparison of the expression profiles of WT and Dnmt3l-KO MEFs. Two biological replicates of WT and Dnmt3l-KO samples were analyzed. The red and green circles indicate genes that were differentially expressed between WT and Dnmt3l-KO MEFs (fold change >1.5 and P value <0.05). FC, fold change; pVal, P value; KO, Dnmt3l-KO. (B) Gene ontology (GO) term analysis of the differentially expressed genes in WT and Dnmt3l-KO MEFs. (C) Bar chart displaying the relative expression levels of differentially expressed genes in WT and Dnmt3l-KO MEFs. The qPCR results were quantified by 2−ΔΔCp method and Gapdh was used as a housekeeping gene for normalization. The error bars represent the s.e.m. from four to five biological repeats. *denotes P<0.05 and **indicates P<0.01 (Student's t test).
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0031
Collectively, these results suggest that the relatively permissive state of chromatin and the increased expression of development-associated genes in Dnmt3l-KO donor cells provide potential benefits for the improvement of NT efficiency.
Discussion
In vitro studies of epigenetic reprogramming hold great potential to improve our knowledge of natural developmental processes and advance techniques for patient-specific therapeutic applications (Nishikawa et al. 2008, Cantone & Fisher 2013, Long et al. 2014). However, successful reprogramming occurs only in a proportion of cloned embryos; therefore, improvements in developmental efficiency and our understanding of epigenetic reprogramming are essential. Our results revealed that a synergistic effect between Dnmt3l-KO donor cells and TSA treatment dramatically increased the efficiency and quality of cloned embryos.
Several reports have shown that TSA treatment of the donor cell or the reconstructed embryo can improve NT cloning efficiency, which is likely through the creation of a permissive chromatin state (Enright et al. 2003, Enright et al. 2005, Kishigami et al. 2006, Rybouchkin et al. 2006, Bui et al. 2010). Recent studies have observed that donor cells with decreased H3K9me3 or H3K27me3 levels are associated with improved NT efficiency in cloned embryos (Antony et al. 2013, Matoba et al. 2014, Saini et al. 2015), further suggesting that the establishment of accessible chromatin can facilitate nuclear reprogramming in cloned embryos. In this study, we observed that Dnmt3l-KO MEFs display relatively higher levels of H3K27ac and lower levels of H3K9me3 and H3K27me3 compared with WT MEFs. Furthermore, cloned embryos generated from Dnmt3l-KO donor cells and treated with TSA significantly increased NT efficiencies at pre-implantation stages. These results suggest that the initial epigenetic state of the donor nucleus contributes to the development of cloned embryos and that the reduction of repressive histone marks in donor cells benefits the reprogramming process.
Genome-wide epigenetic reprogramming occurs at two developmental stages in the mammalian life cycle: gametogenesis and pre-implantation embryonic development (Sasaki & Matsui 2008, Messerschmidt et al. 2014). Dnmt3l/DNMT3L is significantly expressed in both pre-implantation embryo and germ cells (Bourc'his et al. 2001, Sakai et al. 2004, Guenatri et al. 2013). In germ cells, DNMT3L contributes to the monoallelic expression of imprinted genes and TE silencing to safeguard the integrity of the genome for germ cell development (Bourc'his et al. 2001, Bourc'his & Bestor 2004, Hata et al. 2006). DNMT3L also interacts with DNMT3A to establish oocyte-specific methylation imprints during oocyte growth (Lucifero et al. 2007, Hara et al. 2014). Furthermore, recent studies using high-throughput analyses of WT and Dnmt3l-KO oocytes revealed a role for DNMT3L in establishing DNA methylation patterns on transposable elements (Smallwood et al. 2011, Kobayashi et al. 2012, Shirane et al. 2013). In male germ cells, Dnmt3l deficiency leads to a defect in imprinting establishment and silencing of retrotransposons (Bourc'his & Bestor 2004, Hata et al. 2006, Kato et al. 2007, Liao et al. 2014, Vlachogiannis et al. 2015).
In contrast, DNMT3L is not crucial for the modulation of genomic imprinting at pre-implantation stages (Bourc'his et al. 2001, Inoue et al. 2002, Kato et al. 2007, Kamimura et al. 2014). DNMT3L is not strictly required for transposable element silencing in fertilized embryos (Guenatri et al. 2013). Moreover, our results revealed that the repression of TEs, at least MuERVL and IAP, occurs normally in Dnmt3l-KO cloned embryos. These results suggest that either the role of DNMT3L in pre-implantation embryos is different from its function in germ cells or that the remaining Dnmt3l transcripts in the recipient ooplasm are sufficient to regulate MuERVL and IAP expression.
Our previous study indicated that DNMT3L can facilitate interactions among DNMT3A, SETDB1, TRIM28 and HDAC1 in ES cells, suggesting the involvement of DNMT3L in generating repressive chromatin for transcriptional inhibition (Kao et al. 2014). Compared with WT cloned embryos, the increased developmental efficiency of Dnmt3l-KO cloned embryos indicated improved embryonic gene activation in the absence of zygotic Dnmt3l, as we observed significantly increased cell numbers and more restricted OCT4 expression in the ICM at the blastocyst stage. These results suggested that donor MEFs from Dnmt3l-KO embryos possess a more permissive epigenetic environment for nuclear reprogramming. In addition, zygotic Dnmt3l may be dispensable for the function of SETDB1 in ICM establishment, given that disruption of SETDB1 expression results in defective ICM growth in pre-implantation embryos (Dodge et al. 2004); by comparison, Dnmt3l-KO cloned embryos exhibited high-quality blastocysts. These data suggest that zygotic Dnmt3l may not be involved in SETDB1-mediated ICM formation in cloned embryos.
Dnmt3l-KO MEF donor cells displayed increased levels of the active histone mark H3K27ac, reduced nuclear HDAC1 localization and decreased levels of repressive histone marks, suggesting that the chromatin state of these donor cells is more permissive for the reprogramming process. Furthermore, our data revealed that compared with WT MEFs, Dnmt3l-KO MEFs exhibited differential expression of gene associated with embryonic development. The expression of Dnmt3l/DNMT3L was not detected in MEFs by western blotting or quantitative real-time PCR analysis (data not shown); therefore, the observed effects on HDAC1, histone modifications and gene expression may result from accumulation in stem/progenitor cells when DNMT3L is expressed. Relationships between deficiencies of epigenetic regulators and long-term effects at later developmental stages have been previously reported (Bourc'his & Bestor 2004, Gu et al. 2011). For instance, Dnmt3l-KO male germ cells exhibit severe defects at the postnatal spermatocyte stage when Dnmt3l/DNMT3L is not expressed (Bourc'his et al. 2001, Hata et al. 2002). Recently, DNMT3L expression has been observed in several types of stem/progenitor cells (Nimura et al. 2006, Liu et al. 2013, Liao et al. 2014). It will be interesting to investigate how DNMT3L influences the epigenome in stem/progenitor cells and the basis for its potential long-term effects in their descendants.
NT reprogramming in human cell lines has improved in recent years. The recent successful generation of human pluripotent embryonic stem cells (NT-hES cells) provides hope that NT-derived embryos and NT-ESCs may serve as models for human disease research with significant applications for personalized cell replacement therapies (Tachibana et al. 2013, Chung et al. 2014, Ma et al. 2014, Yamada et al. 2014). Accumulating evidence has revealed that DNMT3L/DNMT3L is expressed in several types of human tissues, including the prefrontal cortex, liver, testis, ovary and thymus (Aapola et al. 2000, Borghese et al. 2012, Lee et al. 2014). DNMT3L mutations have been associated with schizophrenia, subtelomeric hypomethylation, lower intelligence and ovarian endometriosis (El-Maarri et al. 2009, Haggarty et al. 2010, Borghese et al. 2012, Saradalekshmi et al. 2014). Recent studies have shown that the loss of Dnmt3l expression correlates with failed TE silencing and defective genomic imprinting at the Dlk1-Dio3 locus in induced pluripotent stem cells (iPSCs) (Stadtfeld et al. 2012, Tang et al. 2012), which might correlate with reduced neural lineage differentiation potential (Stadtfeld et al. 2012, Mo et al. 2015). Dnmt3l-mutant iPSCs are therefore not suitable for biomedical research. As zygotic Dnmt3l is dispensable for blastocyst development and TE silencing in cloned embryo, NT followed by embryonic stem cell derivation is an alternative method and a potential strategy for DNMT3L-related disease research and pharmaceutical development.
In conclusion, our study suggests that the use of Dnmt3l-KO donor cells benefits SCNT reprogramming efficiency, potentially due to a more accessible chromatin state and reduced HDAC1 activity in Dnmt3l-KO donor cells.
Supplementary data
This is linked to the online version of the paper at http://dx.doi.org/10.1530/REP-15-0031.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by grants from the Ministry of Science and Technology, Taiwan (102-2321-B-002-031, 103-2321-B-002-099 to S-P Lin; 103-2811-B-002-098 to H-F Liao) and National Taiwan University (97R0066-46 to S-P Lin; 97R0066-41 to S-C Wu).
References
Aapola U, Kawasaki K, Scott HS, Ollila J, Vihinen M, Heino M, Shintani A, Kawasaki K, Minoshima S & Krohn K et al. 2000 Isolation and initial characterization of a novel zinc finger gene, DNMT3L, on 21q22.3, related to the cytosine-5-methyltransferase 3 gene family. Genomics 65 293–298. (doi:10.1006/geno.2000.6168)
Aapola U, Liiv I & Peterson P 2002 Imprinting regulator DNMT3L is a transcriptional repressor associated with histone deacetylase activity. Nucleic Acids Research 30 3602–3608. (doi:10.1093/nar/gkf474)
Antony J, Oback F, Chamley LW, Oback B & Laible G 2013 Transient JMJD2B-mediated reduction of H3K9me3 levels improves reprogramming of embryonic stem cells into cloned embryos. Molecular and Cellular Biology 33 974–983. (doi:10.1128/MCB.01014-12)
Bock C, Reither S, Mikeska T, Paulsen M, Walter J & Lengauer T 2005 BiQ Analyzer: visualization and quality control for DNA methylation data from bisulfite sequencing. Bioinformatics 21 4067–4068. (doi:10.1093/bioinformatics/bti652)
Borghese B, Santulli P, Hequet D, Pierre G, de Ziegler D, Vaiman D & Chapron C 2012 Genetic polymorphisms of DNMT3L involved in hypermethylation of chromosomal ends are associated with greater risk of developing ovarian endometriosis. American Journal of Pathology 180 1781–1786. (doi:10.1016/j.ajpath.2012.01.009)
Bourc'his D & Bestor TH 2004 Meiotic catastrophe and retrotransposon reactivation in male germ cells lacking Dnmt3L. Nature 431 96–99. (doi:10.1038/nature02886)
Bourc'his D, Xu GL, Lin CS, Bollman B & Bestor TH 2001 Dnmt3L and the establishment of maternal genomic imprints. Science 294 2536–2539. (doi:10.1126/science.1065848)
Bui HT, Wakayama S, Kishigami S, Park KK, Kim JH, Thuan NV & Wakayama T 2010 Effect of trichostatin A on chromatin remodeling, histone modifications, DNA replication, and transcriptional activity in cloned mouse embryos. Biology of Reproduction 83 454–463. (doi:10.1095/biolreprod.109.083337)
Cantone I & Fisher AG 2013 Epigenetic programming and reprogramming during development. Nature Structural & Molecular Biology 20 282–289. (doi:10.1038/nsmb.2489)
Chedin F, Lieber MR & Hsieh CL 2002 The DNA methyltransferase-like protein DNMT3L stimulates de novo methylation by Dnmt3a. PNAS 99 16916–16921. (doi:10.1073/pnas.262443999)
Chung YG, Eum JH, Lee JE, Shim SH, Sepilian V, Hong SW, Lee Y, Treff NR, Choi YH & Kimbrel EA et al. 2014 Human somatic cell nuclear transfer using adult cells. Cell Stem Cell 14 777–780. (doi:10.1016/j.stem.2014.03.015)
Deplus R, Brenner C, Burgers WA, Putmans P, Kouzarides T, de Launoit Y & Fuks F 2002 Dnmt3L is a transcriptional repressor that recruits histone deacetylase. Nucleic Acids Research 30 3831–3838. (doi:10.1093/nar/gkf509)
Ding X, Wang Y, Zhang D, Wang Y, Guo Z & Zhang Y 2008 Increased pre-implantation development of cloned bovine embryos treated with 5-aza-2′-deoxycytidine and trichostatin A. Theriogenology 70 622–630. (doi:10.1016/j.theriogenology.2008.04.042)
Dodge JE, Kang YK, Beppu H, Lei H & Li E 2004 Histone H3-K9 methyltransferase ESET is essential for early development. Molecular and Cellular Biology 24 2478–2486. (doi:10.1128/MCB.24.6.2478-2486.2004)
El-Maarri O, Kareta MS, Mikeska T, Becker T, Diaz-Lacava A, Junen J, Nusgen N, Behne F, Wienker T & Waha A et al. 2009 A systematic search for DNA methyltransferase polymorphisms reveals a rare DNMT3L variant associated with subtelomeric hypomethylation. Human Molecular Genetics 18 1755–1768. (doi:10.1093/hmg/ddp088)
Enright BP, Kubota C, Yang X & Tian XC 2003 Epigenetic characteristics and development of embryos cloned from donor cells treated by trichostatin A or 5-aza-2′-deoxycytidine. Biology of Reproduction 69 896–901. (doi:10.1095/biolreprod.103.017954)
Enright BP, Sung LY, Chang CC, Yang X & Tian XC 2005 Methylation and acetylation characteristics of cloned bovine embryos from donor cells treated with 5-aza-2′-deoxycytidine. Biology of Reproduction 72 944–948. (doi:10.1095/biolreprod.104.033225)
Gu TP, Guo F, Yang H, Wu HP, Xu GF, Liu W, Xie ZG, Shi L, He X & Jin SG et al. 2011 The role of Tet3 DNA dioxygenase in epigenetic reprogramming by oocytes. Nature 477 606–610. (doi:10.1038/nature10443)
Guenatri M, Duffie R, Iranzo J, Fauque P & Bourc'his D 2013 Plasticity in Dnmt3L-dependent and -independent modes of de novo methylation in the developing mouse embryo. Development 140 562–572. (doi:10.1242/dev.089268)
Haggarty P, Hoad G, Harris SE, Starr JM, Fox HC, Deary IJ & Whalley LJ 2010 Human intelligence and polymorphisms in the DNA methyltransferase genes involved in epigenetic marking. PLoS ONE 5 e11329. (doi:10.1371/journal.pone.0011329)
Hara S, Takano T, Fujikawa T, Yamada M, Wakai T, Kono T & Obata Y 2014 Forced expression of DNA methyltransferases during oocyte growth accelerates the establishment of methylation imprints but not functional genomic imprinting. Human Molecular Genetics 23 3853–3864. (doi:10.1093/hmg/ddu100)
Hata K, Okano M, Lei H & Li E 2002 Dnmt3L cooperates with the Dnmt3 family of de novo DNA methyltransferases to establish maternal imprints in mice. Development 129 1983–1993.
Hata K, Kusumi M, Yokomine T, Li E & Sasaki H 2006 Meiotic and epigenetic aberrations in Dnmt3L-deficient male germ cells. Molecular Reproduction and Development 73 116–122. (doi:10.1002/mrd.20387)
Holst CR, Bou-Reslan H, Gore BB, Wong K, Grant D, Chalasani S, Carano RA, Frantz GD, Tessier-Lavigne M & Bolon B et al. 2007 Secreted sulfatases Sulf1 and Sulf2 have overlapping yet essential roles in mouse neonatal survival. PLoS ONE 2 e575. (doi:10.1371/journal.pone.0000575)
Ichiyanagi T, Ichiyanagi K, Miyake M & Sasaki H 2013 Accumulation and loss of asymmetric non-CpG methylation during male germ-cell development. Nucleic Acids Research 41 738–745. (doi:10.1093/nar/gks1117)
Inoue K, Kohda T, Lee J, Ogonuki N, Mochida K, Noguchi Y, Tanemura K, Kaneko-Ishino T, Ishino F & Ogura A 2002 Faithful expression of imprinted genes in cloned mice. Science 295 297. (doi:10.1126/science.295.5553.297)
Jia D, Jurkowska RZ, Zhang X, Jeltsch A & Cheng X 2007 Structure of Dnmt3a bound to Dnmt3L suggests a model for de novo DNA methylation. Nature 449 248–251. (doi:10.1038/nature06146)
Jullien J, Pasque V, Halley-Stott RP, Miyamoto K & Gurdon JB 2011 Mechanisms of nuclear reprogramming by eggs and oocytes: a deterministic process? Nature Reviews. Molecular Cell Biology 12 453–459. (doi:10.1038/nrm3140)
Kamimura S, Hatanaka Y, Hirasawa R, Matsumoto K, Oikawa M, Lee J, Matoba S, Mizutani E, Ogonuki N & Inoue K et al. 2014 Establishment of paternal genomic imprinting in mouse prospermatogonia analyzed by nuclear transfer. Biology of Reproduction 91 120. (doi:10.1095/biolreprod.114.120451)
Kao TH, Liao HF, Wolf D, Tai KY, Chuang CY, Lee HS, Kuo HC, Hata K, Zhang X & Cheng X et al. 2014 Ectopic DNMT3L triggers assembly of a repressive complex for retroviral silencing in somatic cells. Journal of Virology 88 10680–10695. (doi:10.1128/JVI.01176-14)
Kato Y, Kaneda M, Hata K, Kumaki K, Hisano M, Kohara Y, Okano M, Li E, Nozaki M & Sasaki H 2007 Role of the Dnmt3 family in de novo methylation of imprinted and repetitive sequences during male germ cell development in the mouse. Human Molecular Genetics 16 2272–2280. (doi:10.1093/hmg/ddm179)
Kigami D, Minami N, Takayama H & Imai H 2003 MuERV-L is one of the earliest transcribed genes in mouse one-cell embryos. Biology of Reproduction 68 651–654. (doi:10.1095/biolreprod.102.007906)
Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R & Salzberg SL 2013 TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biology 14 R36. (doi:10.1186/gb-2013-14-4-r36)
Kim S, Gunesdogan U, Zylicz JJ, Hackett JA, Cougot D, Bao S, Lee C, Dietmann S, Allen GE & Sengupta R et al. 2014 PRMT5 protects genomic integrity during global DNA demethylation in primordial germ cells and preimplantation embryos. Molecular Cell 56 564–579. (doi:10.1016/j.molcel.2014.10.003)
Kishigami S, Mizutani E, Ohta H, Hikichi T, Thuan NV, Wakayama S, Bui HT & Wakayama T 2006 Significant improvement of mouse cloning technique by treatment with trichostatin A after somatic nuclear transfer. Biochemical and Biophysical Research Communications 340 183–189. (doi:10.1016/j.bbrc.2005.11.164)
Kobayashi H, Sakurai T, Imai M, Takahashi N, Fukuda A, Yayoi O, Sato S, Nakabayashi K, Hata K & Sotomaru Y et al. 2012 Contribution of intragenic DNA methylation in mouse gametic DNA methylomes to establish oocyte-specific heritable marks. PLoS Genetics 8 e1002440. (doi:10.1371/journal.pgen.1002440)
Kurimoto K, Yabuta Y, Ohinata Y & Saitou M 2007 Global single-cell cDNA amplification to provide a template for representative high-density oligonucleotide microarray analysis. Nature Protocols 2 739–752. (doi:10.1038/nprot.2007.79)
Lee SM, Lee YG, Bae JB, Choi JK, Tayama C, Hata K, Yun Y, Seong JK & Kim YJ 2014 HBx induces hypomethylation of distal intragenic CpG islands required for active expression of developmental regulators. PNAS 111 9555–9560. (doi:10.1073/pnas.1400604111)
Liang X, Zhou Q, Li X, Sun Y, Lu M, Dalton N, Ross JJ & Chen J 2005 PINCH1 plays an essential role in early murine embryonic development but is dispensable in ventricular cardiomyocytes. Molecular and Cellular Biology 25 3056–3062. (doi:10.1128/MCB.25.8.3056-3062.2005)
Liao HF, Chen WS, Chen YH, Kao TH, Tseng YT, Lee CY, Chiu YC, Lee PL, Lin QJ & Ching YH et al. 2014 DNMT3L promotes quiescence in postnatal spermatogonial progenitor cells. Development 141 2402–2413. (doi:10.1242/dev.105130)
Liu L, Souto J, Liao WB, Jiang YP, Li YQ, Nishinakamura R, Huang SM, Rosengart T, Yang VW & Schuster M et al. 2013 Histone lysine-specific demethylase 1 (LSD1) protein Is involved in sal-like protein 4 (SALL4)-mediated transcriptional repression in hematopoietic stem cells. Journal of Biological Chemistry 288 34719–34728. (doi:10.1074/jbc.M113.506568)
Long CR, Westhusin ME & Golding MC 2014 Reshaping the transcriptional frontier: epigenetics and somatic cell nuclear transfer. Molecular Reproduction and Development 81 183–193. (doi:10.1002/mrd.22271)
Love MI, Huber W & Anders S 2014 Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology 15 550. (doi:10.1186/s13059-014-0550-8)
Lucifero D, La Salle S, Bourc'his D, Martel J, Bestor TH & Trasler JM 2007 Coordinate regulation of DNA methyltransferase expression during oogenesis. BMC Developmental Biology 7 36. (doi:10.1186/1471-213X-7-36)
Ma P & Schultz RM 2008 Histone deacetylase 1 (HDAC1) regulates histone acetylation, development, and gene expression in preimplantation mouse embryos. Developmental Biology 319 110–120. (doi:10.1016/j.ydbio.2008.04.011)
Ma H, Morey R, O'Neil RC, He Y, Daughtry B, Schultz MD, Hariharan M, Nery JR, Castanon R & Sabatini K et al. 2014 Abnormalities in human pluripotent cells due to reprogramming mechanisms. Nature 511 177–183. (doi:10.1038/nature13551)
Maalouf WE, Liu Z, Brochard V, Renard JP, Debey P, Beaujean N & Zink D 2009 Trichostatin A treatment of cloned mouse embryos improves constitutive heterochromatin remodeling as well as developmental potential to term. BMC Developmental Biology 9 11. (doi:10.1186/1471-213X-9-11)
Mason K, Liu Z, Aguirre-Lavin T & Beaujean N 2012 Chromatin and epigenetic modifications during early mammalian development. Animal Reproduction Science 134 45–55. (doi:10.1016/j.anireprosci.2012.08.010)
Matoba S, Liu Y, Lu F, Iwabuchi KA, Shen L, Inoue A & Zhang Y 2014 Embryonic development following somatic cell nuclear transfer Impeded by persisting histone methylation. Cell 159 884–895. (doi:10.1016/j.cell.2014.09.055)
Messerschmidt DM, Knowles BB & Solter D 2014 DNA methylation dynamics during epigenetic reprogramming in the germline and preimplantation embryos. Genes & development 28 812–828. (doi:10.1101/gad.234294.113)
Mi H, Muruganujan A, Casagrande JT & Thomas PD 2013 Large-scale gene function analysis with the PANTHER classification system. Nature Protocols 8 1551–1566. (doi:10.1038/nprot.2013.092)
Mo CF, Wu FC, Tai KY, Chang WC, Chang KW, Kuo HC, Ho HN, Chen HF & Lin SP 2015 Loss of non-coding RNA expression from the DLK1-DIO3 imprinted locus correlates with reduced neural differentiation potential in human embryonic stem cell lines. Stem Cell Research & Therapy 6 1. (doi:10.1186/scrt535)
Nimura K, Ishida C, Koriyama H, Hata K, Yamanaka S, Li E, Ura K & Kaneda Y 2006 Dnmt3a2 targets endogenous Dnmt3L to ES cell chromatin and induces regional DNA methylation. Genes to Cells 11 1225–1237. (doi:10.1111/j.1365-2443.2006.01012.x)
Nishiyama M, Nakayama K, Tsunematsu R, Tsukiyama T, Kikuchi A & Nakayama KI 2004 Early embryonic death in mice lacking the β-catenin-binding protein duplin. Molecular and Cellular Biology 24 8386–8394. (doi:10.1128/MCB.24.19.8386-8394.2004)
Nishikawa S, Goldstein RA & Nierras CR 2008 The promise of human induced pluripotent stem cells for research and therapy. Nature Reviews. Molecular Cell Biology 9 725–729. (doi:10.1038/nrm2466)
Ogura A, Inoue K & Wakayama T 2013 Recent advancements in cloning by somatic cell nuclear transfer. Philosophical Transactions of the Royal Society of London. Series B, Biological sciences 368 20110329. (doi:10.1098/rstb.2011.0329)
Ooi SK, Qiu C, Bernstein E, Li K, Jia D, Yang Z, Erdjument-Bromage H, Tempst P, Lin SP & Allis CD et al. 2007 DNMT3L connects unmethylated lysine 4 of histone H3 to de novo methylation of DNA. Nature 448 714–717. (doi:10.1038/nature05987)
Pang J, Hoefen R, Pryhuber GS, Wang J, Yin G, White RJ, Xu X, O'Dell MR, Mohan A & Michaloski H et al. 2009 G-protein-coupled receptor kinase interacting protein-1 is required for pulmonary vascular development. Circulation 119 1524–1532. (doi:10.1161/CIRCULATIONAHA.108.823997)
Peaston AE, Evsikov AV, Graber JH, de Vries WN, Holbrook AE, Solter D & Knowles BB 2004 Retrotransposons regulate host genes in mouse oocytes and preimplantation embryos. Developmental Cell 7 597–606. (doi:10.1016/j.devcel.2004.09.004)
Roberts A & Pachter L 2013 Streaming fragment assignment for real-time analysis of sequencing experiments. Nature Methods 10 71–73. (doi:10.1038/nmeth.2251)
Rodriguez-Osorio N, Urrego R, Cibelli JB, Eilertsen K & Memili E 2012 Reprogramming mammalian somatic cells. Theriogenology 78 1869–1886. (doi:10.1016/j.theriogenology.2012.05.030)
Rybouchkin A, Kato Y & Tsunoda Y 2006 Role of histone acetylation in reprogramming of somatic nuclei following nuclear transfer. Biology of Reproduction 74 1083–1089. (doi:10.1095/biolreprod.105.047456)
Saini M, Selokar NL, Raja AK, Sahare AA, Singla SK, Chauhan MS, Manik RS & Palta P 2015 Effect of donor cell type on developmental competence, quality, gene expression, and epigenetic status of interspecies cloned embryos produced using cells from wild buffalo and oocytes from domestic buffalo. Theriogenology 84 101–108e1. (doi:10.1016/j.theriogenology.2015.02.018)
Sakai Y, Suetake I, Shinozaki F, Yamashina S & Tajima S 2004 Co-expression of de novo DNA methyltransferases Dnmt3a2 and Dnmt3L in gonocytes of mouse embryos. Gene Expression Patterns 5 231–237. (doi:10.1016/j.modgep.2004.07.011)
Sambrook J & Russell D 2001 Molecular Cloning: A Laboratory Manual, third edition. New York, USA: Cold Spring Harbor Laboratory Press
Saradalekshmi KR, Neetha NV, Sathyan S, Nair N IV, air CM & Banerjee M 2014 DNA methyl transferase (DNMT) gene polymorphisms could be a primary event in epigenetic susceptibility to schizophrenia. PLoS ONE 9 e98182. (doi:10.1371/journal.pone.0098182)
Sasaki H & Matsui Y 2008 Epigenetic events in mammalian germ-cell development: reprogramming and beyond. Nature Reviews. Genetics 9 129–140. (doi:10.1038/nrg2295)
Schultz DC, Ayyanathan K, Negorev D, Maul GG & Rauscher FJ 2002 SETDB1: a novel KAP-1-associated histone H3, lysine 9-specific methyltransferase that contributes to HP1-mediated silencing of euchromatic genes by KRAB zinc-finger proteins. Genes & Development 16 919–932. (doi:10.1101/gad.973302)
Shirane K, Toh H, Kobayashi H, Miura F, Chiba H, Ito T, Kono T & Sasaki H 2013 Mouse oocyte methylomes at base resolution reveal genome-wide accumulation of non-CpG methylation and role of DNA methyltransferases. PLoS Genetics 9 e1003439. (doi:10.1371/journal.pgen.1003439)
Smallwood SA, Tomizawa S, Krueger F, Ruf N, Carli N, Segonds-Pichon A, Sato S, Hata K, Andrews SR & Kelsey G 2011 Dynamic CpG island methylation landscape in oocytes and preimplantation embryos. Nature Genetics 43 811–814. (doi:10.1038/ng.864)
Stadtfeld M, Apostolou E, Ferrari F, Choi J, Walsh RM, Chen T, Ooi SS, Kim SY, Bestor TH, Shioda T, Park PJ & Hochedlinger K 2012 Ascorbic acid prevents loss of Dlk1-Dio3 imprinting and facilitates generation of all-iPS cell mice from terminally differentiated B cells. Nature Genetics 44 398–405, S391-392. (doi:10.1038/ng.1110)
Suetake I, Shinozaki F, Miyagawa J, Takeshima H & Tajima S 2004 DNMT3L stimulates the DNA methylation activity of Dnmt3a and Dnmt3b through a direct interaction. Journal of Biological Chemistry 279 27816–27823. (doi:10.1074/jbc.M400181200)
Sung LY, Chang CC, Amano T, Lin CJ, Amano M, Treaster SB, Xu J, Chang WF, Nagy ZP & Yang X et al. 2010 Efficient derivation of embryonic stem cells from nuclear transfer and parthenogenetic embryos derived from cryopreserved oocytes. Cellular Reprogramming 12 203–211. (doi:10.1089/cell.2009.0072)
Tachibana M, Amato P, Sparman M, Gutierrez NM, Tippner-Hedges R, Ma H, Kang E, Fulati A, Lee HS & Sritanaudomchai H et al. 2013 Human embryonic stem cells derived by somatic cell nuclear transfer. Cell 153 1228–1238. (doi:10.1016/j.cell.2013.05.006)
Tang Y, Luo Y, Jiang Z, Ma Y, Lin CJ, Kim C, Carter MG, Amano T, Park J & Kish S et al. 2012 Jak/Stat3 signaling promotes somatic cell reprogramming by epigenetic regulation. Stem Cells 30 2645–2656. (doi:10.1002/stem.1225)
Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, Pimentel H, Salzberg SL, Rinn JL & Pachter L 2012 Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks. Nature Protocols 7 562–578. (doi:10.1038/nprot.2012.016)
Vassena R, Dee Schramm R & Latham KE 2005 Species-dependent expression patterns of DNA methyltransferase genes in mammalian oocytes and preimplantation embryos. Molecular Reproduction and Development 72 430–436. (doi:10.1002/mrd.20375)
Vlachogiannis G, Niederhuth CE, Tuna S, Stathopoulou A, Viiri K, de Rooij DG, Jenner RG, Schmitz RJ & Ooi SK 2015 The Dnmt3L ADD domain controls cytosine methylation establishment during spermatogenesis. Cell Reports 10 944–956. (doi:10.1016/j.celrep.2015.01.021)
Yamada M, Johannesson B, Sagi I, Burnett LC, Kort DH, Prosser RW, Paull D, Nestor MW, Freeby M & Greenberg E et al. 2014 Human oocytes reprogram adult somatic nuclei of a type 1 diabetic to diploid pluripotent stem cells. Nature 510 533–536. (doi:10.1038/nature13287)
Yang L, Mei Q, Zielinska-Kwiatkowska A, Matsui Y, Blackburn ML, Benedetti D, Krumm AA, Taborsky GJ Jr & Chansky HA 2003 An ERG (ets-related gene)-associated histone methyltransferase interacts with histone deacetylases 1/2 and transcription co-repressors mSin3A/B. Biochemical Journal 369 651–657. (doi:10.1042/BJ20020854)
Yang X, Smith SL, Tian XC, Lewin HA, Renard JP & Wakayama T 2007 Nuclear reprogramming of cloned embryos and its implications for therapeutic cloning. Nature Genetics 39 295–302. (doi:10.1038/ng1973)
Zhang M, Wang F, Kou Z, Zhang Y & Gao S 2009 Defective chromatin structure in somatic cell cloned mouse embryos. Journal of Biological Chemistry 284 24981–24987. (doi:10.1074/jbc.M109.011973)
Zhou HM, Weskamp G, Chesneau V, Sahin U, Vortkamp A, Horiuchi K, Chiusaroli R, Hahn R, Wilkes D & Fisher P et al. 2004 Essential role for ADAM19 in cardiovascular morphogenesis. Molecular and Cellular Biology 24 96–104. (doi:10.1128/MCB.24.1.96-104.2004)
(H-F Liao and C-F Mo contributed equally to this work)
