bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 1
1 Catalytic inhibition of H3K9me2 writers disturbs epigenetic marks during bovine nuclear
2 reprogramming.
3 1,3,4Sampaio RV*, 1,4Sangalli JR, 1De Bem THC, 1Ambrizi DR, 1 del Collado M, 1 Bridi A, 1 Ávila
4 ACFCM, 2Macabelli CH, 1Oliveira LJ, 1da Silveira JC, 2Chiaratti MR, 1Perecin F,1Bressan FF;
5 3Smith LC, 4Ross PJ, 1Meirelles FV.
6 1 Departament of Veterinary Medicine, Faculty of Animal Science and Food Engineering,
7 University of Sao Paulo, Pirassununga, SP, Brazil. 2Departament of Genetic and Evolution,
8 Federal University of São Carlos, São Carlos, SP, Brazil. 3 Université de Montréal, Faculté de
9 médecine vétérinaire, Centre de recherche en reproduction et fertilité, St. Hyacinthe, Québec,
10 postcode: H3T 1J4, Canada.4Department of Animal Science, University of California Davis,
11 USA.
12 *Correspondence
13 Rafael Vilar Sampaio, Department of Veterinary Medicine, Faculty of Food Engineering and
14 Animal Science, University of Sao Paulo, Pirassununga, Sao Paulo, Brazil. Email:
16 Abstract
17 Orchestrated events, including extensive changes in epigenetic marks, allow a somatic nucleus to
18 become totipotent after transfer into an oocyte, a process termed nuclear reprogramming.
19 Recently, several strategies have been applied in order to improve reprogramming efficiency,
20 mainly focused on removing repressive epigenetic marks such as histone methylation from the
21 somatic nucleus. Herein we used the specific and non-toxic chemical probe UNC0638 to inhibit
22 the catalytic activity of the histone metyltransferases EHMT1 and EHMT2. Either the donor cell
23 (before reconstruction) or the early embryo was exposed to the probe to assess its effect on bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 2
24 developmental rates and epigenetic marks. First, we showed that the treatment of bovine
25 fibroblasts with UNC0638 did mitigate the levels of H3K9me2. Moreover, H3K9me2 levels
26 were decreased in cloned embryos regardless of treating either donor cells or early embryos with
27 UNC0638. Additional epigenetic marks such as H3K9me3, 5mC, and 5hmC were also affected
28 by the UNC0638 treatment. Therefore, the use of UNC0638 did diminish the levels of H3K9me2
29 and H3K9me3 in SCNT-derived blastocysts, but this was unable to improve their
30 preimplantation development. These results indicate that the specific reduction of H3K9me2 by
31 inhibiting EHMT1/2 causes diverse modifications to the chromatin during early development,
32 suggesting an intense epigenetic crosstalk during nuclear reprogramming.
33 Keywords: SCNT; cattle; chromatin remodeling; histone-lysine methyltransferase
34
35 INTRODUCTION
36 Cloning by somatic cell nuclear transfer (SCNT) is an inefficient technique largely because of
37 incomplete nuclear reprogramming. The persistent epigenetic memory from the somatic cell
38 nucleus is considered one of the main barriers for an efficient reprogramming process 1. In most
39 cases, the recipient oocyte used in SCNT fails to erase residual epigenetic marks from somatic
40 nucleus, which are retained in the embryo and lead to abnormal gene expression 2. For instance,
41 when compared to in vitro fertilized (IVF) embryos, SCNT embryos show an aberrant pattern of
42 gene expression during the period of embryonic genome activation (EGA) 3,4. According to
43 previous works, aberrant gene expression in cloned embryos is linked with altered DNA and
44 histone demethylation 5,6. Hence, epigenetic errors might lead to problems such as abortion,
45 placental and fetal abnormalities, among others 7. bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 3
46 One of the main anomalies identified during SCNT development relates to its aberrant
47 EGA 3,4,8. Recently, genomic areas refractory to reprogramming were identified and named as
48 “reprogramming resistant regions” (RRR) 8. In cattle, major EGA occurs at the 8/16-cell stage 9
49 and it is possible that augmented levels of H3K9me2 and H3K9me3 during this stage are
50 responsible for the poor development of SCNT-derived embryos in cattle 6,10.
51 Modification on the lysine residue at the position 9 on histone H3 is one of the most
52 studied epigenetic marks. EHMT2 and the related molecule EHMT1 (also known as G9a and
53 GLP) are the main enzymes responsible for methylation at this site 11. EHMT1/2 are SET
54 (Su(var)3-9, E(z), and Trithorax) domain-containing proteins, a conserved domain present in
55 other epigenetic enzymes that uses the cofactor S-adenosyl-l-methionine (SAM) to achieve
56 methylation on histones and other proteins 12. They also have an Ankyrin (ANK) repeat domain,
57 which can generate and read the same epigenetic mark and are considered as a motif to interact
58 with other proteins 13. There is a well-known crosstalk between DNA methylation and histone
59 modifications. DNA methyltransferase DNMT1 directly binds EHMT2 to coordinate their
60 modification during DNA replication 14. EHMT2 is also implicated in genomic imprinting
61 regulation 15,16 and its absence results in the impairment of placenta-specific imprinting 17.
62 However, the enzymatic activity of EHMT1/2 has recently been shown to be dispensable for
63 imprinted DNA methylation, since EHMT1/2 can recruit DNMTs via its ANK domain 18.
64 Small molecules have been used to inhibit EHMT1/2 in an attempt to reduce H3K9me2
65 and facilitate nuclear reprogramming during induced pluripotent stem cell (iPSC) generation 19,20
66 and SCNT 21,22; however, even with the intense crosstalk between histone modification and DNA
67 methylation, the effects of the inhibition of H3K9me2 writers on DNA methylation and other
68 epigenetic marks remain unknown. In this study, we hypothesized that the inhibition of bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 4
69 EHMT1/2 activity either before or after nuclear results in decreased H3K9me2 levels and
70 improves development of SCNT-derived embryos. Additionally, we investigated throughout
71 preimplantation development the consequences of EHMT1/2 inhibition on additional epigenetic
72 marks and expression of genes related to epigenetic regulation.
73
74 RESULTS
75 EHMT1/2 inhibition reduces H3K9me2 levels in donor cells
76 To test whether the UNC0638 treatment would be suitable for the bovine species, fetal
77 fibroblasts were treated with 250 nM of UNC0638 for 96 h and the levels of H3K9me2 were
78 analyzed by immunofluorescence (IF) and western blot (WB). Both IF and WB results showed
79 that UNC0638 treatment decreased H3K9me2 levels in bovine fibroblasts when compared to
80 non-treated controls (Fig.1A-D). Moreover, the treatment with UNC0638 did not impact on the
81 levels of H3K9me3, validating the efficiency and specificity of UNC0638 as a bovine EHMT1/2
82 inhibitor.
83 Next, we aimed at investigating whether the lower levels of H3K9me2 were kept after
84 nuclear transfer. Hence, these fibroblasts were used as nuclear donors. In agreement with the
85 result found in treated fibroblasts (Figure 1), the use of these cells as nuclear donors revealed that
86 lower levels of H3K9me2 were present in 1-cell embryos 18 h after oocyte reconstruction. This
87 gives evidence that the chromatin remodeling occurring immediately after nuclear transfer was
88 not able to revert the treatment effect on H3K9me2 levels (Figure 2). To further evaluate the
89 implications of the treatment, we also assessed the levels of H3K9me3, 5mC and 5hmC in 1-cell
90 embryos. Although the levels of 5mC and 5hmC remained unchanged, increased levels of
91 H3K9me3 were found after 18 h of oocyte reconstruction (Figure 2). Given that the levels of bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 5
92 H3K9me3 were unaltered in fibroblasts prior to nuclear transfer (Figure 1), it is likely that this
93 change occurred after oocyte reconstruction.
94 Similar to what was seen in 1-cell embryos, the levels of H3K9m2 were decreased in
95 both 8/16-cell embryos and blastocysts (Figure 3 and 4) produced using donor cells treated with
96 UNC0638. This result supports the idea that the low levels of H3K9m2 presented in fibroblasts
97 were kept after nuclear transfer and up to the blastocyst stage. In contrast to the 1-cell stage, the
98 levels of H3K9me3 were no longer altered in 8/16-cell embryos derived from UNC0638 treated
99 cells (Figure 3). Furthermore, lower levels of H3K9me3 were found in these embryo at the
100 blastocyst stage (Figure 4). With respect to 5mC and 5hmC, the levels of these marks were
101 increased in 8/16-cell stage embryos reconstructed using UNC0638-treated cells (Figure 3), but
102 not altered in blastocysts (Figure 4). This indicates a complex and dynamic epigenetic effect of
103 the UNC0638 treatment on DNA methylation during early development.
104 In cattle, the high levels of H3K9me2 present in somatic nuclei persist during
105 development of SCNT-derived embryos and might play a role as a barrier for nuclear
106 reprogramming 6,23. Given the success of the UNC0638 treatment in decreasing the levels of
107 H3K9me2 in fibroblasts and SCNT-derived embryos, we assessed its impact on developmental
108 rates. Remarkably, no effect was observed on both cleavage (p = 0.33) and blastocyst (p = 0.73)
109 rates (Table 1). Therefore, our findings suggest that the levels of H3K9me2 present in donor
110 cells are not linked with the preimplantation development of SCNT embryos.
111
112 EHMT1/2 inhibition in SCNT embryos affects epigenetic marks during early development
113 In agreement with the above findings, culture of cloned embryos for 55 h in the presence
114 of UNC0638 (from oocyte reconstruction to the 8/16-cell stage) also led to decreased levels of bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 6
115 H3K9me2 and increased levels of H3K9me3, 5mC and 5hmC in 8/16-cell stage embryos (Figure
116 5). In this experiment donor cells were not treated, providing evidence that the treatment after
117 oocyte reconstruction was also effective. Moreover, the levels of both H3K9me2 and H3K9me3
118 were decreased at the blastocyst stage in comparison with untreated embryos (Figure 6). This
119 reversal of H3K9me3 from higher to lower levels is consistent with that observed when the
120 treatment was performed in donor cells (Figure 2 to 4). Similarly, decreased levels of 5hmC were
121 present in blastocysts derived from the UNC0638 treatment (Figure 6), suggesting that
122 UNC0638 results in dynamic changes on the levels of H3K9me3, 5mC and 5hmC (Figure 3).
123 Overall, in spite of slight differences, treatment of either donor cells or early embryos with
124 UNC0638 resulted in a similar pattern of epigenetic modification.
125 Next, we sought to determine the effect of the UNC0638 treatment on transcript levels of
126 i) the main inhibited targets, such as EHMT1 and EHMT2; ii) genes potentially related to the
127 activity of EHMTs, such as DNMT1, DNMT3A, and DNMT3B 14,24; and, iii) genes linked with
128 the role played by H3K9me2 during embryonic development, such as TET1, TET2, and TET3 25.
129 However, no effect on transcript abundance was found when considering 8/16-cell embryos and
130 blastocysts (Figure 7). These results indicate that the changes observed on epigenetic marks were
131 independent of gene expression, i.e. at a post-transcriptional level.
132 Finally, treatment of SCNT-derived embryos with UNC0638 did not impact on both
133 cleavage (p = 0.65) and blastocyst (p = 0.13) rates (Table 2).
134
135 DISCUSSION
136 In summary, our findings indicate that the treatment with UNC0638 was able to reduce
137 the levels of H3K9me2 regardless of treating either donor cells or the early embryo derived from bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 7
138 SCNT. Importantly, the lower levels of H3K9me2 were maintained throughout the
139 preimplantation development, but this failed to enhance developmental rates. Additionally, the
140 levels of H3K9me3, 5mC and 5hmC in SCNT-derived embryos were impacted by the UNC0638
141 treatment, but not the transcriptional activity of genes related to the role of EHMTs and
142 H3K9me2.
143 Residual epigenetic memory has been widely considered responsible for improper
144 reprogramming of somatic cells 26. The high levels of H3K9me2 present in somatic nuclei persist
145 during the development of bovine SCNT embryos 6,23. Aiming to decrease H3K9me2 levels, we
146 used UNC0638, a small molecule that specifically inhibits the catalytic activity of histone
147 methyltransferases EHMT1/2 in a non-toxic manner as previously shown in mice and sheeps
148 27,28. Likewise, our treatment proved effective, resulting in lower levels of H3K9me2 in somatic
149 cells exposed to UNC0638. This reduction was observed throughout embryogenesis of SCNT
150 embryos, regardless of the moment of the treatment (i.e. in donor cells or in early embryos). We
151 then hypothesized that distinct time of expression between H3K9me2 writers and erasers might
152 explain the lower levels of H3K9me2 since the expression of EHMT1 and EHMT2 in bovine is
153 higher in oocytes than in early embryos 29 while the expression of lysine demethylases is
154 increased later in development 30. Herein, we found no effect of the UNC0678 treatment when
155 main regulatory genes were analyzed (Figure 7). This result is consistent with a previous report
156 with somatic cells in which the UNC0638 treatment did not affect neither the protein levels of
157 EHMT1 or EHMT2 nor the mRNA levels of EHMT2, indicating that changes in H3K9me2 were
158 a consequence of decreased enzymatic activity 28.
159 As expected, different epigenetic marks were also found to be changed herein by
160 EHMT1/2 inhibition. During the earliest stages of embryogenesis, we observed an increase in bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 8
161 H3K9me3 levels, which was followed by a drop in blastocysts. This increase in H3K9me3 has
162 been reported as a classical epigenetic alteration secondary to the low levels of H3K9me2 due
163 the preferential activity of SUV39H for mono or di-methylated H3K9 as a substrate to form
164 H3K9me3 31,32. In fact, demand for substrate along with demethylase activity is a reasonable
165 explanation for the subsequent decline in H3K9me3 levels as H3K9me2 is required for
166 H3K9me3 formation. In agreement with this, treatment with an EHMT2 inhibitor led to a
167 decrease in H3K9me3 at target genes 33. Moreover, common genes were upregulated by the
168 knockdown of EHMT2 or SUV39H1 34, suggesting they share similar targets. However, given
169 that H3K9me3 was not changed in somatic cells, these dynamic changes in H3K9me3 levels
170 likely require methylation writers and erasers present in the early embryo 35.
171 Regarding DNA modifications, analysis of SCNT embryos revealed an increase at 8/16-
172 cell stage in 5mC and 5hmC levels, regardless of the moment the treatment was performed
173 (Figure 2-4). Due to the fact that 5mC is required for 5hmC formation 36,37, we speculate that the
174 increase in 5hmC was likely a consequence of increased 5mC availability. Actually, we expected
175 an opposite result given that knockout of EHMT1/2 in ESCs leads to global DNA demethylation
176 38,39. This is in agreement with the idea that the presence of H3K9me2 in the maternal
177 pronucleus, together with DPPA3/STELLA, prevents active DNA demethylation by Tet-family
178 enzymes 40. Nonetheless, it was recently shown, both in IVF and SCNT embryos, that DNA
179 demethylation may be independent of H3K9me2; only DPPA3/STELLA was needed 41.
180 Moreover, a previous work showed that the treatment with an EHMT2 inhibitor not only
181 decreased H3K9me2, but also increased DNA methylation in neuroblastoma cells 42. It might be
182 possible that herein inhibition of EHMT1/2 catalytic activity prompted EHMT1/2 to cooperate in
183 additional functions. In fact, the involvement of EHMT2 in DNA methylation goes beyond its bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 9
184 methylase activity 24,38. However, the process involved in this offset mechanism remains elusive.
185 The DNA hypermethylation of cloned embryos is one of the major failures of proper chromatin
186 remodeling during nuclear reprogramming 43,44, however, in contrast to changes at a global level,
187 severe demethylation in important regions as satellite I and imprinted genes 43,44 could also be
188 responsible for such failure.
189 Finally, its noteworthy that the unaltered transcript levels of UNC0638-treated embryos
190 suggest a very specific effect of the treatment on EHMT1/2 inhibition. This is consistent with the
191 finding that treatment with UNC0638 does not affect developmental rates as reported here and
192 previously in mice 45, goats 46, and sheep 27,47. It is likely the effect of the UNC0638 is species-
193 specific as only in pigs it was shown to improve development of SCNT-derived embryos 21,48.
194 Likewise, one might expect the UNC0638 treatment to improve developmental rates as it
195 significantly mitigated the levels of H3K9me3 in blastocysts; this mark is considered a main
196 epigenetic barrier for nuclear reprogramming. First described in mice, the RRRs are H3K9me3-
197 enriched regions refractory to transcriptional activation in 2-cell cloned embryos, the moment
198 when EGA occurs in mice 8. Similarly, reduction of H3K9me3 levels during bovine EGA
199 improved SCNT efficiency 10, indicating that in this species, 8/16-cells stage may be the best
200 period to test H3K9me3 reduction-based approaches. Taken together, these data suggest that
201 EHMT1/2 contribute to the stability of different epigenetic marks during bovine reprogramming
202 independently of their catalytic activity.
203
204 CONCLUSION
205 Using a specific probe to inhibit the catalytic activity of EHMT1/2, we were able to
206 mitigate the levels of H3K9me2 throughout early embryonic development. Inhibition, either in bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 10
207 nuclear donor cells or during early development of SCNT-derived embryos triggered changes in
208 H3K9me3, 5mC, and 5hmC levels independently of transcriptional changes. This suggests an
209 intense crosstalk among histone modifications and DNA methylation during nuclear
210 reprogramming in cattle. Nonetheless, although lower levels of H3K9me2 and H3K9me3 were
211 achieved at the blastocyst stage, EHMT1/2 inhibition did not increase developmental rates.
212 Therefore, the aberrant epigenetic remodeling present in SCNT-derived embryos may not be
213 amended by inhibiting the catalytic activity of these epigenetic writers.
214
215 METHODS
216 All chemicals and reagents used were purchased from Sigma-Aldrich Chemical Company (St.
217 Louis, MO, USA), unless otherwise stated. The present study was approved by the ethical
218 committee for the use of animals of the School of Veterinary Medicine and Animal Science of
219 the University of Sao Paulo, protocol number 9076280114, and in agreement with the ethical
220 principles of animal research. We adopted the International Guiding Principles for Biomedical
221 Research Involving Animals (Society for the Study of Reproduction) as well.
222
223 Cell culture. Fibroblast cell line was obtained from a 55-days-old crossbred (Gir x Holstein; Bos
224 indicus x Bos taurus) fetus, as described previously 49. Concisely, a skin biopsy (1 cm2) was
225 minced into small pieces with a scalpel blade, followed by digestion with 0.1% (w/v) collagenase
226 for 3 h at 38.5 °C. The digested product was centrifuged at 300 × g for 5 min, and the pellet was
227 resuspended in culture medium (α-MEM [GIBCO BRL, Grand Island, NY, USA] supplemented
228 with 10% [v/v] fetal bovine serum [FBS] and 50 μg/ml gentamicin sulfate). Then, plated in 35-
229 mm Petri dishes and cultured for 6 days in an incubator to establish the primary culture. After bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 11
230 achievement of 70% of confluence (approximately 4x105 cells), cells were detached from the
231 plate using Tryple Express (GIBCO BRL); resuspended in culture medium supplemented with
232 10% of dimethyl sulfoxide (DMSO), and 50 μg/mL gentamicin sulfate; placed in cryotubes; and
233 stored in liquid N2 until use.
234
235 Cell treatment with UNC0638. For nuclear donor treatment, primary fetal fibroblasts were
236 thawed and cultured as described above. After 24 h, culture medium was replaced with fresh
237 medium containing 250nM of UNC0638, best concentration described previously 28. Forty-eight
238 hours later, UNC0638 treatment was renewed for an additional 48h under serum starvation
239 (culture media supplemented with 0.5% of FBS instead of 10%) to arrest the cell cycle at the
240 G1/G0 stages. Dimethyl sulfoxide (DMSO) at same concentration of UNC0638 dilution (0.01%)
241 and submitted to same culture conditions was used as control. Nighty-six hours after starting
242 treatment (48h + 48h), cells were then used for immunostaining, western blotting analysis or
243 SCNT. While for embryo treatment, nuclear donors were cultured for 48h under starvation
244 medium without treatment.
245 Immunostaining of somatic cells. For H3K9me2 and H3K9me3 analysis, we procedure as
246 described previously 50,51 with few modifications. Cells were plated on coverslips and treated as
247 described above. Then, cells were fixed with 4% paraformaldehyde for 15 min and
248 permeabilized with D-PBS + 1% Triton X-100 for 10min and blocked for 30 min in D-PBS +
249 0.3%Triton X-100 + 1% BSA, were placed in primary antibody solution consisting of blocking
250 buffer, a mouse antibody anti-H3K9me2 - Abcam (ab1220, 1:300) and a rabbit antibody anti-
251 H3K9me3 - Abcam (ab8898, 1:500), overnight at 4 °C. Cells incubated without primary
252 antibodies were used as negative controls for all assays. In the next day, after washing 3x for 10 bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 12
253 min each, cells were incubated with secondary antibodies Alexa Fluor 568-conjugated goat anti-
254 rabbit IgG (Life Tech, cat. # A-11036 and Alexa Fluor 488-conjugate goat anti-mouse IgG (Life
255 Tech, cat. #: A-11029) both at RT for 1h. Then, after washing for 3x for 10 min each, samples
256 were mounted on microscope slides and with Prolong Gold Antifade Mountant (Life Tech, cat. #
257 P36935). Images were captured using a confocal microscope (TCS-SP5 AOBS; Leica, Soims,
258 Germany) using laser excitation and emission filters specific for Alexa 488 and Alexa 568.
259 Digital images were analyzed by evaluating each nucleus fluorescent intensity using ImageJ-Fiji
260 image processing software (National Institutes of Health, Bethesda, MD). The fluorescent
261 intensity (average mean gray value) of each channel were measured by manually outlining each
262 nucleus and adjusted against background.
263
264 Protein analysis. All steps for histone acid extraction and western blot were described in details
265 previously, with few modifications 49.
266 Briefly, histones were extracted using the EpiQuick Total Histone Extraction Kit (Epigentek,
267 Farmingdale, NY, USA, cat. # OP-0006) following the manufacturer’s instructions with few
268 modifications. After harvested, cells were centrifuged at 300 xg for 5 min at 4°C, and the pellet
269 were incubated in 200μL of pre-lysis buffer, and centrifuged again at 9,500 xg for 1 min at 4 °C.
270 Then, the supernatant was removed and the pellet was resuspended in 60μL of lysis buffer. After
271 incubation for 30 min at 4 °C, they were centrifuged at 13,500 xg for 5 min at 4 °C. The
272 containing acid-soluble proteins present in the supernatant was transferred into a new vial and
273 neutralized with 0.3 volumes of balance buffer supplemented with DTT. The proteins extracted
274 were quantified by Qubit 2.0 (Life Technologies) and stored in aliquots at −20 °C.
275 bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 13
276 Western Blotting for quantification of H3K9me2 levels. To measure the relative H3K9me2
277 levels, 0.5μg of histones were mixed with 4x Laemmli buffer (Bio Rad Laboratories, Hercules,
278 CA, USA cat. #161-0747), following by denaturation at 98 °C for 5 min, and loaded into an
279 acrylamide gel. Proteins were fractionated by size on a 4%–15% SDS-PAGE gel run at a
280 constant 100 V for 90 min. Then, proteins were electroblotted using a semi-dry transfer system
281 (Trans Blot turbo Bio Rad), onto PVDF membranes (Bio-Rad) at a constant 25 V for 3 min. The
282 membranes were blocked with 5% BSA in Tris-buffered saline (TBS) + 0.1% of TWEEN (TBS-
283 T) for 1 h at room temperature, following by incubation overnight at 4 °C under agitation with
284 the primary antibody anti-H3K9me2 (Abcam ab1220) diluted 1:5,000 in TBS-T + 1% BSA
285 solution. In the next day, the membranes were washed 3x in TBS-T for 5 min each and incubated
286 for 1 h with peroxidase-conjugated anti-rabbit secondary antibody (Sigma, cat # A0545) diluted
287 1:10,000 in TBS-T + 1% BSA solution. The membranes were incubated with Clarity Western
288 ECL Substrate (Bio Rad cat. # 170-5060) for 30 seconds, and images were captured using the
289 ChemiDoc MP Imaging System (Bio-Rad). The analysis of images and band quantification were
290 made using the software Image Lab 5.1 (Bio-Rad). The loading control histone H3 was used to
291 calculate the abundance of H3K9me2. To this end, the same amount of proteins was run in
292 parallel, then the membranes were incubated overnight at 4 °C under agitation with anti-histone
293 H3 antibody (Sigma, cat. # H0164) diluted 1:10,000 in TBS-T + 1% BSA solution, then washed
294 and developed as described above.
295
296 Somatic cell nuclear transfer. SCNT was performed as described 52, with few modifications.
297 Bovine ovaries were obtained from a local slaughterhouse and transported to the laboratory in
298 saline solution at 38.5°C. Using a syringe and 18G needle from 3 to 6 mm antral follicles, bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 14
299 cumulus oocyte complexes (COCs) were aspirated and cultured in Tissue Culture Medium 199
300 (TCM199) supplemented with 10% FBS, 0.2 mM pyruvate, 50 µg/mL gentamicin, 0.5 mg/mL
301 FSH, and hCG, 5 U/mL (Vetecor, Hertape Calier) for 18h under mineral oil at 38.5°C and in
302 atmosphere with 5% CO2. Then, to remove cumulus cells from oocytes, manual pipetting was
303 made in the presence of hyaluronidase (0.2%). Only oocytes with evident first polar body (PB)
304 were selected for enucleation. To visualize the DNA oocytes were labeled with 10 µg/mL of the
305 DNA fluorochrome Hoechst 33342 for 15 minutes at RT in SOF supplemented with 5 mg/mL
306 BSA, 2.5% FCS, 0.2 mM sodium pyruvate, gentamicin 53. Then, they were washed and
307 transferred to a micromanipulation drop of Medium 199, Hanks' Balanced Salts plus Hepes
308 (Gibco, Grand Island, US, CAT 12350-039) supplemented 10% FBS, 0.2 mM pyruvate, 50
309 µg/mL gentamicin, and 7.5 µg/ mL cytochalasin B, under mineral oil. All procedures were
310 performed on an inverted microscope (Nikon Eclipse Ti) equipped with Hoffman optics and
311 Narishige micromanipulators (Tokyo, Japan). A 15-um inner diameter enucleation pipette (ES
312 TransferTip; Eppendorf) was used to aspirate and remove the PB and metaphase II plates. The
313 aspirated cytoplasm was exposed to UV light to confirm for the presence of PB and metaphase
314 plate. After enucleation, one single donor cell from control or treatment, at the same cell passage,
315 was placed into the perivitelline space of each enucleated oocyte. To fuse them the reconstructed
316 embryos were placed into a fusion chamber filled with 0.3 M mannitol in a Multiporator
317 (Eppendorf, Hamburg, Germany) with settings of one pulse of alternating current (5 seconds,
318 50V/cm) followed by two continuous current electric pulses (45 µs each, 1.75 kV/cm). Then,
319 they were incubated for at least 60 min to select the successfully fused. After 26 h post-
320 maturation, the fused SCNT couplets were activated with 5 µM ionomycin in TCM199-HEPES
321 (supplemented with 1 mg/mL fatty acid-free BSA) for 5 minutes and moved into TCM199- bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 15
322 HEPES medium (supplemented with 30 mg/mL BSA) for 3 minutes. Then, cultured for 3h in
323 SOF medium supplemented with 6-dimethylaminopurine. Following activation, presumptive
324 zygotes from cell treatment were washed and placed into droplet of 100 µL SOF medium, while
325 for the embryo treatment groups, zygotes from non-treated cells were split into drops of control
326 (SOF medium supplemented with 0.01% of DMSO) or treatment (SOF medium supplemented
327 with 250nM of UNC0638 until reach 8/16 cells stage 55h) placed in different plates under
328 mineral oil at 38.5°C and in atmosphere with 5% CO2. Developmental rates were evaluated at
329 day 2 for cleavage and day 7 for blastocyst. Control for oocyte quality, activation procedure, and
330 in vitro culture was made using activated metaphase II-arrested oocytes, using the same
331 activation protocol described above (parthenogenetic embryos). Embryos at one cell stage (18
332 hours post-activation), 8/16-cells stage (55 hours post-activation), and at blastocyst stage were
333 collected either for immunostaining procedure or for gene expression analysis.
334
335 Embryo immunostaining. Immunostaining was performed as described previously 50, with
336 modifications. Embryos at 1/2-cell stage (18h post activation), 8/16-cells stage (55h post
337 activation), and blastocyst at day 7 after activation were collected and washed in PBS+PVA.
338 Then, embryos were fixed with 4% paraformaldehyde for 15 min and permeabilized with D-PBS
339 + 1% Triton X-100 for 30min. Exclusively for 5mC and 5hmC staining, following
340 permeabilization step, we treated embryos with 4N HCl for 15 min. Next, we proceed with
341 neutralization for 20 min with 100 mM Tris-HCl buffer (pH 8.5). After blocking altogether for 2
342 h in D-PBS + 0.1%Triton X-100 + 1% BSA, embryos were placed in primary antibody solution
343 consisting of blocking buffer, a mouse antibody anti-H3K9me2 - Abcam (ab1220, 1:300), a
344 rabbit antibody anti-H3K9me3 - Abcam (ab8898, 1:500), a mouse antibody anti-5-mC - Abcam bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 16
345 (ab10805, 1:500), and a rabbit antibody anti-5-hydroxymethylcytosine (5-hmC) antibody -
346 Abcam (ab214728, 1:500) overnight at 4 °C. After washing 3x for 10 min and 3x for 20 min
347 each, embryos were incubated with secondary antibodies Alexa Fluor 568-conjugated goat anti-
348 rabbit IgG (Life Tech, cat. # A-11036 and Alexa Fluor 488-conjugate goat anti-mouse IgG (Life
349 Tech, cat. #: A-11029) both at RT for 1h. Then, after washing for 3x for 10 min and 3x for 20
350 min each, embryos were mounted on slides with Prolong Gold Antifade Mountant (Life Tech,
351 cat. # P36935). Z-stack images were captured using a confocal microscope (TCS-SP5 AOBS;
352 Leica, Soims, Germany) using laser excitation and emission filters specific for Alexa 488 and
353 Alexa 568. Pictures were false-colored using LAS Lite Leica software for informative purpose
354 according to each epigenetic mark (H3K9me2-green, H3K9me3-red, 5mC-cyan, and 5hmC-
355 magenta); no further adjustments or image processing was done. Digital images were analyzed
356 by evaluating each nucleus fluorescent intensity using ImageJ-Fiji image processing software
357 (National Institutes of Health, Bethesda, MD). After maximum projection reconstruction of Z-
358 stacks, the fluorescent intensity (average mean gray value) of each channel were measured by
359 manually outlining each nucleus and adjusted against cytoplasmic background. At least 10
360 embryos at each stage from 3 different replicates were analyzed.
361
362 RNA extraction, reverse transcription and quantification. Total RNA of 8/16-cell embryos
363 (n= 5) and blastocysts (n =5) from 3 biological replicates were extracted using TRIzol Reagent
364 (InvitrogenTM, Carlsbad, CA) and Linear Acrylamide (InvitrogenTM, Carlsbad, CA), according
365 to the manufacturer’s instruction. Quality and quantification of RNA were performed using
366 NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific; Absorbance 260/280nm ratio).
367 Next, RNA was treated with Deoxyribonuclease I, Amplification Grade (DNaseI, Invitrogen; bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 17
368 Carlsbad, CA) for 15min at room temperature to digest any contaminating DNA. DNase I was
369 inactivated by adding 1 L of EDTA in the reaction for 10 min at 65°C. Double-stranded cDNA
370 from 8/16-cell embryos and blastocysts were synthesized from 160ng of total RNA with final
371 volume of 20µl using High Capacity cDNA Reverse transcription Kit (Thermo Fisher). The
372 reaction was composed by total RNA, 10x RT Buffer, 10x RT Random primers, 25x dNTP MIX,
373 nuclease-free water and Reverse Transcriptase. After, the reactions were incubated at 25°C for
374 10 minutes, 37°C for 120 minutes and 85°C for 5 minutes. Quantitative real-time polymerase
375 chain reactions (qRT-PCR) were performed in QuantStudio 6 Flex (Applied Biosystems)
376 equipment using a final volume of 10 µL per well with 2x Power SYBR Green RT-PCR Master
377 Mix (Applied Biosystems), 1 µM/mL of forward and reverse bovine-specific primers, 6.95ng of
378 cDNA and nuclease-free water. The primers were designed using the software Primer-BLAST
379 (NCBI) based upon sequences available in GenBank (Supplemental Table 1). The primers were
380 also sequenced to test their specificity. DNTM1, DNMT3A, DNMT3B and the reference genes
381 PPIA and RPL15 were used as described previously 54,55. Amplification were performed with
382 initial denaturation at 95°C for 2 minutes, following by 45 cycles at 95°C for 15 seconds and
383 60°C for 1 minute, followed by melting curve analysis to verify the presence of a single
384 amplified product. The mRNAs were considered present when the cycle threshold (CT) was less
385 than 37 cycles and presented adequate melting curve. Samples of mature oocytes, 8-16 cells
386 embryos and blastocysts were analyzed in duplicate and the CT data was normalized using
387 geometric mean of PPIA and RPL15 reference genes.
388
389 Statistical analysis. Statistical analysis was performed using GraphPad Prism 7 (GraphPad
390 Software, San Diego, California, USA). Data were tested for normality of residuals and bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 18
391 homogeneity of variances using the Shapiro-Wilk test and analyzed as follows. Immunostaining
392 and western blot data were analyzed by Student’s t-test. Gene expression experiments data were
393 analyzed by one-way ANOVA followed by Tukey’s post hoc test. Frequency data
394 (developmental rates) were analyzed by Student’s t-test. Differences with probabilities of P <
395 0.05 were considered significant. In the text, values are presented as the means ± the standard
396 error of the mean (S.E.M.). All experiments were repeated at least three times unless otherwise
397 stated.
398 Data Availability. The datasets generated during and/or analyzed during the current study are
399 available from the corresponding author on reasonable request.
400
401 Author Contributions. R.V.S. and F.V.M. conceptualized the study; R.V.S. collected the
402 samples; R.V.S., J.R.S., T.H.C.B, D.R.A., M.C., A.B., A.C.F.C.F.A, and C.H.M. performed the
403 experiments; R.V.S., J.C.S., F.P., and M.R.C performed formal data analysis and interpretation;
404 F.V.M., F.F.B., L.J.O, L.C.S, and P.J.R manage and coordinate research planning and execution,
405 and oversight the research activity; R.V.S., L.C.S., M.R.C., and F.V.M. wrote the manuscript
406 with input of all authors. All the authors discussed the results, critically reviewed and edited the
407 manuscript.
408
409 Acknowledgment
410 The authors would like to thank the staff and students at the LMMD, Jessica Brunhara Cruz,
411 João Vitor Puttini Paixão, and Rodrigo Barreto for their assistance with the sample collections,
412 laboratory procedures and discussions. The slaughterhouses Olhos D`água and Vale do Prata for
413 gently provided ovaries for this study. Rafael Vilar Sampaio is supported by São Paulo Research bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 19
414 Foundation - FAPESP, grant number #2015/08807-6 and previously by the grants #2015/25111-
415 5 and #2013/07160-3. This work was granted by São Paulo Research Foundation – FAPESP,
416 grant number #2012/50533-2, #2013/08135-2, #2014/21034-3, #2014/22887-0 and
417 #2014/50947-7; National Counsel of Technological and Scientific Development (CNPq) grant
418 number 465539/2014-9 and CAPES. The funders had no role in study design, data collection and
419 analysis, decision to publish, or preparation of the manuscript.
420 Reference
421 1. Pasque, V., Jullien, J., Miyamoto, K., Halley-Stott, R. P. & Gurdon, J. B. Epigenetic
422 factors influencing resistance to nuclear reprogramming. Trends Genet. 27, 516–25
423 (2011).
424 2. Ng, R. K. & Gurdon, J. B. Epigenetic memory of active gene transcription is inherited
425 through somatic cell nuclear transfer. Proc. Natl. Acad. Sci. U. S. A. 102, 1957–62 (2005).
426 3. Suzuki, T., Minami, N., Kono, T. & Imai, H. Zygotically activated genes are suppressed in
427 mouse nuclear transferred embryos. Cloning Stem Cells 8, 295–304 (2006).
428 4. Vassena, R. et al. Tough beginnings: alterations in the transcriptome of cloned embryos
429 during the first two cell cycles. Dev. Biol. 304, 75–89 (2007).
430 5. Dean, W. et al. Conservation of methylation reprogramming in mammalian development:
431 aberrant reprogramming in cloned embryos. Proc. Natl. Acad. Sci. U. S. A. 98, 13734–8
432 (2001).
433 6. Santos, F. et al. Epigenetic Marking Correlates with Developmental Potential in Cloned
434 Bovine Preimplantation Embryos. Curr. Biol. 13, 1116–1121 (2003).
435 7. Smith, L. C. et al. Developmental and epigenetic anomalies in cloned cattle. Reprod.
436 Domest. Anim. 47 Suppl 4, 107–14 (2012). bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 20
437 8. Matoba, S. et al. Embryonic development following somatic cell nuclear transfer impeded
438 by persisting histone methylation. Cell 159, 884–895 (2014).
439 9. Graf, A. et al. Fine mapping of genome activation in bovine embryos by RNA sequencing.
440 Proc. Natl. Acad. Sci. U. S. A. 111, 4139–44 (2014).
441 10. Liu, X. et al. H3K9 demethylase KDM4E is an epigenetic regulator for bovine embryonic
442 development and a defective factor for nuclear reprogramming. Development 145,
443 dev158261 (2018).
444 11. Tachibana, M. et al. G9a histone methyltransferase plays a dominant role in euchromatic
445 histone H3 lysine 9 methylation and is essential for early embryogenesis. Genes Dev. 16,
446 1779–91 (2002).
447 12. Herz, H.-M., Garruss, A. & Shilatifard, A. SET for life: biochemical activities and
448 biological functions of SET domain-containing proteins. Trends Biochem. Sci. 38, 621–
449 639 (2013).
450 13. Collins, R. E. et al. The ankyrin repeats of G9a and GLP histone methyltransferases are
451 mono- and dimethyllysine binding modules. Nat. Struct. Mol. Biol. 15, 245–250 (2008).
452 14. Estève, P.-O. et al. Direct interaction between DNMT1 and G9a coordinates DNA and
453 histone methylation during replication. Genes Dev. 20, 3089–103 (2006).
454 15. Hanna, C. W. & Kelsey, G. Genomic imprinting beyond DNA methylation: a role for
455 maternal histones. Genome Biol. 18, 177 (2017).
456 16. Weaver, J. R. & Bartolomei, M. S. Chromatin regulators of genomic imprinting. Biochim.
457 Biophys. Acta - Gene Regul. Mech. 1839, 169–177 (2014).
458 17. Wagschal, A. et al. G9a histone methyltransferase contributes to imprinting in the mouse
459 placenta. Mol. Cell. Biol. 28, 1104–13 (2008). bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 21
460 18. Zhang, T. et al. G9a/GLP Complex Maintains Imprinted DNA Methylation in Embryonic
461 Stem Cells. Cell Rep. 15, 77–85 (2016).
462 19. Shi, Y. et al. A combined chemical and genetic approach for the generation of induced
463 pluripotent stem cells. Cell Stem Cell 2, 525–8 (2008).
464 20. Shi, Y. et al. Induction of pluripotent stem cells from mouse embryonic fibroblasts by
465 Oct4 and Klf4 with small-molecule compounds. Cell Stem Cell 3, 568–74 (2008).
466 21. Cao, Z. et al. TSA and BIX-01294 Induced Normal DNA and Histone Methylation and
467 Increased Protein Expression in Porcine Somatic Cell Nuclear Transfer Embryos. (2017).
468 doi:10.1371/journal.pone.0169092
469 22. Wu, J. et al. The landscape of accessible chromatin in mammalian preimplantation
470 embryos. Nature 534, 652–657 (2016).
471 23. Wu, X. et al. Multiple histone site epigenetic modifications in nuclear transfer and in vitro
472 fertilized bovine embryos. Zygote 19, 31–45 (2011).
473 24. Epsztejn-Litman, S. et al. De novo DNA methylation promoted by G9a prevents
474 reprogramming of embryonically silenced genes. Nat. Struct. Mol. Biol. 15, 1176–83
475 (2008).
476 25. Kang, J., Kalantry, S. & Rao, A. PGC7, H3K9me2 and Tet3: regulators of DNA
477 methylation in zygotes. Cell Res. 1–4 (2012). doi:10.1038/cr.2012.117
478 26. Pasque, V., Jullien, J., Miyamoto, K., Halley-Stott, R. P. & Gurdon, J. B. Epigenetic
479 factors influencing resistance to nuclear reprogramming. Trends Genet. 27, 516–25
480 (2011).
481 27. Fu, L., Yan, F.-X., An, X.-R. & Hou, J. Effects of the Histone Methyltransferase Inhibitor
482 UNC0638 on Histone H3K9 Dimethylation of Cultured Ovine Somatic Cells and bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 22
483 Development of Resulting Early Cloned Embryos. Reprod. Domest. Anim. 49, e21–e25
484 (2014).
485 28. Vedadi, M. et al. A chemical probe selectively inhibits G9a and GLP methyltransferase
486 activity in cells. Nat. Chem. Biol. 7, 648 (2011).
487 29. McGraw, S., Vigneault, C. & Sirard, M.-A. Temporal expression of factors involved in
488 chromatin remodeling and in gene regulation during early bovine in vitro embryo
489 development. Reproduction 133, 597–608 (2007).
490 30. Glanzner, W. G. et al. Histone 3 lysine 4, 9, and 27 demethylases expression profile in
491 fertilized and cloned bovine and porcine embryos†. Biol. Reprod. (2018).
492 doi:10.1093/biolre/ioy054
493 31. Peters, A. H. F. M. et al. Partitioning and plasticity of repressive histone methylation
494 states in mammalian chromatin. Mol. Cell 12, 1577–89 (2003).
495 32. Rea, S. et al. Regulation of chromatin structure by site-specific histone H3
496 methyltransferases. Nature 406, 593–599 (2000).
497 33. Kubicek, S. et al. Reversal of H3K9me2 by a small-molecule inhibitor for the G9a histone
498 methyltransferase. Mol. Cell 25, 473–481 (2007).
499 34. Kondo, Y. et al. Downregulation of histone H3 lysine 9 methyltransferase G9a induces
500 centrosome disruption and chromosome instability in cancer cells. PLoS One 3, e2037
501 (2008).
502 35. Ross, P. J. & Sampaio, R. V. Epigenetic remodeling in preimplantation embryos: cows are
503 not big mice. Anim. Reprod. 15, 204–214 (2018).
504 36. Iqbal, K., Jin, S., Pfeifer, G. P. & Szabó, P. E. Reprogramming of the paternal genome
505 upon fertilization involves genome-wide oxidation of 5-methylcytosine. 2–7 (2011). bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 23
506 doi:10.1073/pnas.1014033108/-
507 /DCSupplemental.www.pnas.org/cgi/doi/10.1073/pnas.1014033108
508 37. Wossidlo, M. et al. 5-Hydroxymethylcytosine in the mammalian zygote is linked with
509 epigenetic reprogramming. Nat. Commun. 2, 241 (2011).
510 38. Dong, K. B. et al. DNA methylation in ES cells requires the lysine methyltransferase G9a
511 but not its catalytic activity. EMBO J. 27, 2691–2701 (2008).
512 39. Xin, Z. et al. Role of histone methyltransferase G9a in CpG methylation of the Prader-
513 Willi syndrome imprinting center. J. Biol. Chem. 278, 14996–15000 (2003).
514 40. Nakamura, T. et al. PGC7 binds histone H3K9me2 to protect against conversion of 5mC
515 to 5hmC in early embryos. Nature 486, 2–7 (2012).
516 41. Wang, Q.-Q. et al. Dimethylated histone H3 lysine 9 is dispensable for the interaction
517 between developmental pluripotency-associated protein 3 (Dppa3) and ten-eleven
518 translocation 3 (Tet3) in somatic cells. Reprod. Fertil. Dev. 31, 347 (2018).
519 42. Lu, Z. et al. Histone-lysine methyltransferase EHMT2 is involved in proliferation,
520 apoptosis, cell invasion, and DNA methylation of human neuroblastoma cells. Anticancer.
521 Drugs 24, 484–493 (2013).
522 43. Silveira, M. M. et al. DNA methylation profile at a satellite region is associated with
523 aberrant placentation in cloned calves. Placenta 70, 25–33 (2018).
524 44. Suzuki, J. et al. Loss of methylation at H19 DMD is associated with biallelic expression
525 and reduced development in cattle derived by somatic cell nuclear transfer. Biol. Reprod.
526 84, 947–56 (2011).
527 45. Terashita, Y. et al. Latrunculin A Treatment Prevents Abnormal Chromosome
528 Segregation for Successful Development of Cloned Embryos. PLoS One 8, e78380 bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 24
529 (2013).
530 46. Wang, Y. et al. Treatment donor cells with UNC0638 modify the abnormal histone H3K9
531 dimethylation and gene expression in cloned goat embryos. Small Rumin. Res. 156, 27–32
532 (2017).
533 47. Fu, L. et al. Abnormal histone H3K9 dimethylation but normal dimethyltransferase
534 EHMT2 expression in cloned sheep embryos. Theriogenology 78, 1929–38 (2012).
535 48. Wang, F. et al. BIX-01294 increases pig cloning efficiency by improving epigenetic
536 reprogramming of somatic cell nuclei. REPRODUCTION 151, 39–49 (2015).
537 49. Sangalli, J. R. et al. Metabolic gene expression and epigenetic effects of the ketone body
538 β-hydroxybutyrate on H3K9ac in bovine cells, oocytes and embryos. Sci. Rep. 8, 13766
539 (2018).
540 50. Bakhtari, A. & Ross, P. J. DPPA3 prevents cytosine hydroxymethylation of the maternal
541 pronucleus and is required for normal development in bovine embryos. Epigenetics 9,
542 1271–1279 (2014).
543 51. Sampaio, R. V, Chiaratti, M. R., Santos, D. C. N. & Bressan, F. F. Generation of bovine (
544 Bos indicus ) and buffalo ( Bubalus bubalis ) adipose tissue derived stem cells : isolation
545 , characterization , and multipotentiality. Genet. Mol. Res. 14, 53–62 (2015).
546 52. Sá, A. L. et al. Effect of POU5F1 Expression Level in Clonal Subpopulations of Bovine
547 Fibroblasts Used as Nuclear Donors for Somatic Cell Nuclear Transfer. Cell. Reprogram.
548 19, (2017).
549 53. Wells, D. N. Production of Cloned Calves Following Nuclear Transfer with Cultured
550 Adult Mural Granulosa Cells. Biol. Reprod. 60, 996–1005 (1999).
551 54. Macabelli, C. H. et al. Reference gene selection for gene expression analysis of oocytes bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 25
552 collected from dairy cattle and buffaloes during winter and summer. PLoS One 9, e93287
553 (2014).
554 55. Perecin, F. [UNESP]. Epigenética do desenvolvimento em bovinos: DNA
555 metiltransferases e genes imprinted em embriões, fetos e placentas. Aleph xx, 83 f. : il.
556 (2007).
557
558 Figure Legends
559 Figure 1. UNC0638 reduces the abundance of the H3K9me2 in bovine fibroblasts. (A) Western
560 blots and (B) quantification of H3K9me2 of fetal fibroblasts treated with either 0.01% of DMSO
561 (CONTROL) or 250 nM of UNC0638 (UNC) for 96 h. Results from three pairs of biological
562 replicates are shown. H3K9me2 levels were calculated in relation to total histone 3 (H3).
563 Western blot images were cropped for illustrative purposes and images of full blots are presented
564 in Supplemental Fig.1. (C) Levels of H3K9me2 and H3K9me3 determined by
565 immunofluorescence (IF). Representative images of bovine fetal fibroblasts treated for 96 h with
566 0.01% DMSO (Control) or 250nM UNC0638 (UNC0638). Images show fibroblasts nuclei
567 stained with DAPI (left column), anti-H3K9me2 (red; middle column), and anti-H3K9me3
568 (green; right column) antibodies. Images were taken using the same magnification (63x) and
569 laser power, thereby enabling direct comparison of signal intensities. (D) Quantification of
570 H3K9me2 and H3K9me3 fluorescence intensity in nuclei of fibroblasts treated or not with
571 UNC0638 (UNC). Values are presented as mean ± S.E.M. and different letters indicate
572 significant differences (p<0.05).
573 Figure 2 Levels of H3K9me2, H3K9me3, 5mC, and 5hmC in SCNT 1-cell embryos derived
574 from nuclear donor cells treated with UNC0638. (A) One-cell embryos derived from control bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 26
575 donor cells are displayed on first lane and zygotes derived from donor cells treated with
576 UNC0638 are displayed on second lane. Columns 1, 2, 4, and 5 show 1-cell embryos stained
577 with antibody anti-H3K9me2 (green), anti-H3K9me3 (red), anti-5mC (cyan), and anti-5hmC
578 (magenta), respectively. Merge images from H3K9me2+H3K9me3 and 5mC+5hmC are
579 displayed in columns 3 and 6, respectively. All images were taken at the same magnification and
580 at the same laser power, thereby enabling direct comparison of signal intensities. Scale bar 100
581 µm. (B) Quantification of H3K9me2, H3K9me3, 5mC, and 5hmC levels in SCNT embryos
582 derived from donor cells treated or not with UNC0638. Embryos nuclei from 3 different
583 biological replicates were analyzed to investigate the levels of H3K9me2 + H3K9me3 (NT-
584 Control N=13; NT-UNC N=10) and 5mC + 5hmC (NT-Control N=9; NT-UNC0638 N=10).
585 Data are presented as mean ± S.E.M. Different letters above bars indicate significantly
586 differences (p<0.05). NT-Control: Zygotes derived from cells treated with 0.01% of DMSO. NT-
587 UNC: Zygotes derived from cells treated with 250nM of UNC0638.
588 Figure 3 Levels of H3K9me2, H3K9me3, 5mC, and 5hmC in 8/16-cells embryos derived from
589 nuclear donors previous treated with UNC0638. (A) Embryos at 8/16-cells stage derived from
590 nuclear donors control are displayed on first lane and embryos derived from cells treated with
591 UNC0638 are displayed on second lane. Columns 1, 2, 4, and 5 are showing embryos stained
592 with antibody anti-H3K9me2 (green), anti-H3K9me3 (red), anti-5mC (cyan), and anti-5hmC
593 (magenta) are displayed in columns 3 and 6, respectively. All images were taken at the same
594 magnification and at the same laser power, thereby enabling direct comparison of signal
595 intensities. Scale bar 100 µm. (B) Quantification of H3K9me2, H3K9me3, 5mC, and 5hmC
596 levels in SCNT embryos derived from cells treated or not with UNC0638. Embryos nuclei from
597 3 different biological replicates were analyzed to investigate the levels of H3K9me2 + H3K9me3 bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 27
598 (NT-Control N=10; NT-UNC N=8) and 5mC + 5hmC (NT-Control N=8; NT-UNC0638 N=9).
599 Data as presented as mean ± S.E.M, and bars with different letters are significantly different
600 p<0.05). NT-Control: 8/16-cells embryos derived from cells treated with 0.01% of DMSO. NT-
601 UNC: 8/16-cells embryos derived from cells treated with 250nM of UNC0638.
602 Figure 4 Levels of H3K9me2, H3K9me3, 5mC, and 5hmC in blastocysts derived from nuclear
603 donor cells treated with UNC0638. (A) Blastocysts derived from nuclear donor controls are
604 displayed in first lane and embryos derived from cells treated with UNC0638 are displayed in
605 second lane. Columns 1, 2, 4, and 5 are showing embryos stained with antibody anti-H3K9me2
606 (green), anti-H3K9me3 (red), anti-5mC (cyan), and anti-5hmC (magenta), respectively. Merge
607 images from H3K9me2+H3K9me3 and 5mC+5hmC are displayed in columns 3 and 6,
608 respectively. All images were taken at the same magnification and at the same laser power,
609 thereby enabling direct comparison of signal intensities. Scale bar 100 µm. (B) Quantification of
610 H3K9me2, H3K9me3, 5mC, and 5hmC levels in SCNT embryos derived from cells treated or
611 not with UNC0638. Embryos nuclei from 3 different biological replicates were analyzed to
612 investigate the levels of H3K9me2 + H3K9me3 (NT-Control N=15; NT-UNC N=12) and 5mC +
613 5hmC (NT-Control N=11; NT-UNC0638 N=13). Data as presented as the mean ± S.E.M. and
614 means with different letters are significantly different (p<0.05). NT-Control: Blastocysts derived
615 from cells treated with 0.01% of DMSO. NT-UNC: Blastocyst derived from cells treated with
616 250nM of UNC0638.
617 Figure 5 Levels of H3K9me2, H3K9me3, 5mC, and 5hmC in 8/16-cells embryos treated with
618 UNC0638. (A) Embryos at 8/16-cell stage controls are displayed in first lane and embryos
619 previous treated with UNC0638 are displayed in second lane. Columns 1, 2, 4, and 5 are showing
620 embryos stained with antibody anti-H3K9me2 (green), anti-H3K9me3 (red), anti-5mC (cyan), bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 28
621 and anti-5hmC (magenta), respectively. Merge images from H3K9me2+H3K9me3 and
622 5mC+5hmC are displayed in columns 3 and 6, respectively. All images were taken at the same
623 magnification and at the same laser power, thereby enabling direct comparison of signal
624 intensities. Scale bar 100 µm. (B) Quantification of H3K9me2, H3K9me3, 5mC, and 5hmC
625 levels in SCNT embryos treated or not with UNC0638. Embryos nuclei from 3 different
626 biological replicates were analyzed to investigate the levels of H3K9me2 + H3K9me3 (NT-
627 Control N=8; NT-UNC N=8) and 5mC + 5hmC (NT-Control N=8; NT-UNC0638 N=7). Data
628 are presented as mean ± S.E.M. Bars with different letters are significantly different (p<0.05).
629 NT-Control: 8/16-cells embryos previous treated with 0.01% of DMSO. NT-UNC: 8/16-cell
630 embryos previous treated with 250nM of UNC0638.
631 Figure 6 Levels of H3K9me2, H3K9me3, 5mC, and 5hmC in blastocysts treated with
632 UNC0638. (A) Embryos at 8/16-cell stage controls are displayed in first lane and embryos
633 previous treated with UNC0638 are displayed in second lane. Columns 1, 2, 4, and 5 are showing
634 embryos stained with antibody anti-H3K9me2 (green), anti-H3K9me3 (red), anti-5mC (cyan),
635 and anti-5hmC (magenta), respectively. Merge images from H3K9me2+H3K9me3 and
636 5mC+5hmC are displayed in columns 3 and 6, respectively. All images were taken at the same
637 magnification and at the same laser power, thereby enabling direct comparison of signal
638 intensities. Scale bar 100 µm. (B) Quantification of H3K9me2, H3K9me3, 5mC, and 5hmC
639 levels in SCNT embryos treated or not with UNC0638. Embryos nuclei from 3 different
640 biological replicates were analyzed to investigate the levels of H3K9me2 + H3K9me3 (NT-
641 Control N=10; NT-UNC N=7) and 5mC + 5hmC (NT-Control N=7; NT-UNC0638 N=8). Data
642 are presented the mean ± S.E.M. Bars with different letters are significantly different (p<0.05). bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 29
643 NT-Control: Blastocysts derived from embryos previous treated with 0.01% of DMSO. NT-
644 UNC: Blastocysts derived from embryos previous treated with 250nM of UNC0638.
645 Figure 7. Treatment with UNC0638 SCNT embryos during early development does not change
646 transcript levels of epigenetic writer and eraser genes at 8/16-cell and blastocyst stages. The
647 mRNA levels of EHMT1, EHMT2, DNMT1, DNMT3A, DNMT3B, TET1, TET2, and TET3 were
648 quantified using real-time RT-PCR of SCNT embryos treated or not with 250nM of UNC0638.
649 Pool of 5 embryos from 3 different biological replicates (total 15 per group) were analyzed to
650 investigate the levels. Transcript quantities were normalized to the geometric mean of two
651 housekeeping genes (PPIA and RPL15). Data are shown as relative expression, and the values as
652 the mean ± S.E.M. (p > 0.05).
653 654 Supplemental Data Legends 655 656 Supplemental Figure 1 Full immunoblots from data presented on Figure 1. (A)-Total histone 3. 657 1 (B) Levels of H3K9me2 658
659 Supplemental table 1. Primer sequences used in Real Time qPCR.
660 Accession number Primer Sequences (5’-3’) Product (bp) AT (°C)
661
662 Table 1. Development of SCNT zygotes using nuclear donor cells treated with the EHMT1/2
663 inhibitor UNC0638 before oocyte reconstruction.
Treatment Cultured zygotes Cleavage ± S.E.M % Blastocyst % ± S.E.M
NT-CONTROL 287 70.89 ± 5.6 26.95 ± 5.5 NT-UNC 300 62.18 ± 6.6 24.31 ± 5.2 664 bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 30
665 Note: Development rates were obtained from the number of embryos that cleaved or reached the
666 blastocyst stage per reconstructed oocytes (cultured zygotes). Results from 9 replicates.
667 Table 2. Development of SCNT-reconstructed oocytes treated with the EHMT1/2 inhibitor
668 UNC0638 from activation to the 8/16-cell stage.
Treatment Cultured zygotes Cleavage ± S.E.M % Blastocyst % ± S.E.M
NT-CONTROL 249 83.76 ± 2.1 29.34 ± 3.4 NT-UNC 255 82.06 ± 3.0 35.83 ± 2.3 669
670 Note: Development rates were obtained from the number of embryos that cleaved or reached the
671 blastocyst stage per reconstructed oocytes placed in culture (cultured zygotes). Results from 10
672 replicates.
673 Figure 1.
A C DNA H3K9me2 H3K9me3
H3 total
Control Control Control
DNA H3K9me2 H3K9me3 H3K9me2
1 -1 2 2 3 -3 - 8 - 8- - 8 L 3 L 3 L 3 O 6 O 6 O 6 R 0 R 0 R 0 T C T C T C N N N N N N O U O U O U C C C UNC0638 UNC0638 UNC0638 8 B D
0.5 a p=0.044 H3K9me2 H3K9me3 a p<0.001 60 80 a p=0.1040 0.4 a
b 60 0.3 40 b 40 0.2 20 20 Average Intensity Average 0.1 Intensity Average
0 0
Relative level (H3K9me2/H3) level Relative 0.0 O L C OL C CONTROLS UNC0638N CONTROLSO UNC0638UN CONTROL UNC0638NC M O U M L D R DR O U T T R N NT N O O O 674 C C C bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 31
675 Figure 2
A H3K9me2 H3K9me3H3K9me3 Merge 5mC 5hmC Merge
TN-ControlNT-Control
TN-UNCNT-UNC B H3K9me2 H3K9me3 5mC 5hmC 15000 p=0.0018 200000 p=0.0050 100000 a p=0.7032 a 100000 p=0.8794 b a 80000 80000 a a 150000 10000 a 60000 60000 100000 40000 40000 5000 b 50000 20000
20000 Relative intensinty Relative intensinty Relative intensinty Relative intensinty
0 0 0 0 L C L L C L C C O N O N O N O N R R -U R U R U T -U T T T - T - T N N T N T N N O N N N O O O C -C -C -C - T T T T 676 N N N N
677 Figure 3
A H3K9me2 H3K9me3 Merge 5mC 5hmC Merge
NT-Control
NT-UNC
B
H3K9me2 H3K9me3 5mC 5hmC p=0.0244 p=0.1502 p=0.0017 p=0.0135 50000 80000 a 80000 30000 a a b b 40000 b 60000 60000 a 20000 a 30000 40000 40000 20000 10000 20000 20000 10000 Relative intensinty Relative intensinty Relative intensinty Relative intensinty
0 0 0 0 L C L C L C L C O N O N O N O N R -U R -U R -U R -U T T T T T T T T N N N N N O N O N O N O -C -C -C -C T T T T 678 N N N N bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 32
679 Figure 4
A H3K9me2 H3K9me3 Merge 5mC 5hmC Merge
NT-Control
NT-UNC
B H3K9me2 H3K9me3 5mC 5hmC 200 150 80 100 p<0.001 p<0.001 aap=0.9033 p=0.1483 a a a a 80 150 60 100 60 b b 100 40 40 50 50 20 20 Relative intensinty intensinty Relative Relative intensinty intensinty Relative Relative intensinty intensinty Relative Relative intensinty intensinty Relative
0 0 0 0 L C O N L C L C L C R O N O N O N T -U R R U R T T -U T - T -U N N T T T O N N N N N N -C O O O T -C -C -C N T T T 680 N N N
681 Figure 5
A H3K9me2 H3K9me3 Merge 5mC 5hmC Merge
NT-Control
NT-UNC
B H3K9me2 H3K9me3 5mC 5hmC 30000 P<0.05 80000 P<0.05 b 150 p<0.001 150 p<0.001 a b 60000 b b 20000 a 100 100 40000 a 10000 50 a 50 a 20000 Relative intensinty Relative intensinty intensinty Relative Relative intensinty Relative intensinty 0 0 0 0 L C L C L L C O N O N C O N R U R O N R U T - -U R U T - N T T T T - N T N N N N T N O O N O -C C O -C T - -C T N T T N 682 N N bioRxiv preprint doi: https://doi.org/10.1101/847210; this version posted November 20, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 33
683 Figure 6
A IVF
H3K9me2 H3K9me3 Merge 5mC 5hmC Merge
NT-Control
NT-UNC
B H3K9me2 H3K9me3 5mC 5hmC 150000 P<0.05 100000 P<0.05 150 P=0.91 200 P<0.05 a 80000 b aa a a 150 100000 b 100 60000 b 100 40000 50000 50 50 20000 Relative intensinty intensinty Relative Relative intensinty intensinty Relative Relative intensinty Relative intensinty
0 0 0 0 C L C L C L C O N O N R R T -U T -U N T N T ONTROL NT-UN ONTRO NT-UN O N O N -C -C -C -C T T T T 684 N N N N
685
686 Figure 7
DNMT3B ns EHMT1 DNMT1 ns TET2 0.20 0.25 ns 0.08 0.020 ns NT-CONTROL NT-CONTROL NT-CONTROL NT-CONTROL 0.20 0.15 0.06 0.015 NT-UNC 0.15 NT-UNC NT-UNC NT-UNC 0.10 0.04 ns 0.010 0.10 ns 0.05 0.02 0.005 0.05 ns Relative expression Relative Relative expression Relative expression Relative Relative expression Relative ns
0.00 0.00 0.00 0.000 8-16 cells Blastocysts 8-16 cells Blastocysts 8-16 cells Blastocysts 8-16 cells Blastocysts
EHMT2 DNMT3A TET1 TET3 0.10 0.6 ns 0.05 0.5 ns ns ns NT-CONTROL NT-CONTROL NT-CONTROL NT-CONTROL 0.08 0.04 0.4
NT-UNC 0.4 NT-UNC NT-UNC NT-UNC 0.06 ns 0.03 ns 0.3
0.04 0.02 0.2 0.2 ns 0.02 0.01 0.1 Relative expression Relative Relative expression Relative expression Relative Relative expression Relative ns 0.00 0.0 0.00 0.0 687 8-16 cells Blastocysts 8-16 cells Blastocysts 8-16 cells Blastocysts 8-16 cells Blastocysts