Premature Chromosome Condensation Is Not Essential for Nuclear Reprogramming in Bovine Somatic Cell Nuclear Transfer1

Abstract Premature chromosome condensation (PCC) was believed to promote nuclear reprogramming and to facilitate cloning by somatic cell nuclear transfer (NT) in mammalian species. However, it is still uncertain whether PCC is necessary for the successful reprogramming of an introduced donor nucleus in cattle. In the present study, fused NT embryos were subjected to immediate activation (IA, simultaneous fusion and activation), delayed activation (DA, activation applied 4 h postfusion), and IA with aged oocytes (IAA, activation at the same oocyte age as group DA). The morphologic changes, such as nuclear swelling, the occurrence of PCC, and microtubule/aster formation, were analyzed in detail by laser-scanning confocal microscopy. When embryos were subjected to IA in both IA and IAA groups, the introduced nucleus gradually became swollen, and a pronuclear-like structure formed within the oocyte, but PCC was not observed. In contrast, delaying embryo activation resulted in 46.5%–91.2% of NT embryos exhibiting PCC. This PCC was observed beginning at 4 h postcell fusion and was shown as one, two, or multiple chromosomal complexes. Subsequently, a diversity of pronuclear-like structures existed in NT embryos, characterized as single, double, and multiple nuclei. In the oocytes exhibiting PCC, the assembled spindle structure was observed to be an interactive mass, closely associated with condensed chromosomes, but no aster had formed. Regardless of whether they were subjected to IA, IAA, or DA treatments, if the oocytes contained pronuclear-like structures, either one or two asters were observed in proximity to the nuclei. A significantly higher rate of development to blastocysts was achieved in embryos that were immediately activated (IA, 59.1%; IAA, 40.7%) than in those for which activation was delayed (14.2%). The development rate was higher in group IA than in group IAA, but it was not significant (P = 0.089). Following embryo transfer, there was no statistically significant difference in the pregnancy rates (Day 70) between two of the groups (group IA, 11.7%, n = 94 vs. group DA, 12.3%, n = 130; P > 0.05) or live term development (group IA, 4.3% vs. group DA, 4.6%; P > 0.05). Our study has demonstrated that the IA of bovine NT embryos results in embryos with increased competence for preimplantational development. Moreover, PCC was shown to be unnecessary for the reprogramming of a transplanted somatic genome in a cattle oocyte.


INTRODUCTION
1 Somatic cell nuclear transfer (NT) has successfully produced live clones in several 2 mammalian species. In most NT studies, a highly differentiated somatic nucleus is 3 transferred into a recipient oocyte at metaphase II, where nuclear modification and 4 reprogramming take place [1][2][3][4]. During the several hours of exposure to the MII 5 cytoplast, prior to parthenogenetic activation, the introduced somatic nucleus (at the G0 6 and/or G1 phase) usually undergoes nuclear envelope breakdown (NEBD) and 7 subsequent premature chromosome condensation (PCC) [5], likely due to the high 8 concentrations of maturation promoting factor (MPF) present in the oocyte [6,7]. 9 The degree of PCC varies, depending upon the MPF activity and the duration that a 10 transplanted nucleus is exposured to the MII cytoplast [2,5]. In most cloning studies in 11 mice [8, 9] and cattle [10][11][12], parthenogenetic activation was delayed for 1 to 4 h after 12 nuclear transplantation. This was believed to allow extensive nuclear-oocyte interaction, 13 under the hypothesis that a longer exposure of the somatic donor nucleus to the oocyte 14 iodide (P-4170) to stain for DNA, and observed under laser-scanning confocal 1 microscopy (Leica TCS SP2; Mannheim, Germany). 2

Sampling of Reconstructed Bovine Embryos and Morphological Evaluation 3
This study consisted of three experimental treatments that were defined as 4 immediate activation (IA), also called simultaneous fusion and activation; delayed 5 activation (DA); or immediate activation with aged oocytes (IAA). Enucleation (removal 6 of the MII plate) began at 21 h post oocyte maturation. In the IA and DA groups, cell 7 fusion was performed at 25 h post maturation. The time of completion of cell fusion (25  8 hpm was designated as 0 h, and marked as the onset for sample collection). In Group IA,9 immediately after cell fusion (activation at 25 hpm), restructured oocytes were allocated 10 to the activation regime of 5 h incubation in combined CD/CHX and CHX medium, as 11 described above. With this activation procedure, MPF activity in the oocyte decreases to 12 a basal level within 1 h post activation, and the oocyte is dramatically driven away from 13 MII phase into, presumably, the S-phase [21]. In Group DA, fused oocytes were 14 incubated in M199-FBS for 4 h after cell fusion prior to parthenogenetic activation 15 (activation at 29 hpm); subsequently, they were allocated to the same activation regime as 16 Group IA. This incubation period allowed for extended exposure of the introduced 17 G0/G1 nucleus in the oocyte's cytoplasm that contained a high level of MPF to induce 18 PCC. In Group IAA, oocytes were enucleated at 21 hpm. Enucleated cytoplasts were 19 subsequently cultured for 4 h before being subjected to donor cell insertion, cell fusion 20 and a simultaneous activation. Therefore, Groups DA and IAA were treated with a 21 regime that eliminates the age difference between two groups at the time of activation 22 (Groups DA and IAA, 29 hpm). In Groups IA and IAA, the activation treatments were the same except for oocyte age difference at activation (Group IA, 25 hpm vs. Group 1 IAA, 29 hpm). In all three groups, reconstructed oocytes were collected and fixed at 0, 1, 2 2, 4, 6, 12, 18, or 24 h post fusion (hpf). The samples were also collected from three 3 groups at 44 hpf to determine the events of mitosis of NT embryos. The progressive 4 changes over time of the donor nucleus in the oocyte were categorized as: enlarged 5 nucleus or swelling, PCC, and the arrangement of microtubules/asters; these parameters 6 were examined by confocal microscopy. To determine nuclear change, the nuclear area 7 of each reconstructed embryo was analyzed by the Leica Confocal Software program 8 (Leica TCS SP2; Mannheim, Germany). The degree of nuclear swelling was determined 9 by comparison to the average area value of cumulus nuclei from donor cells prior to NT. 10 The evaluation of nuclear swelling was subjectively defined. Those nuclei with areas 11 below (<100%) or similar (100-120% ) to that average value determined for cumulus 12 nuclei were categorized as unswollen; while those with areas 120 % or greater than the 13 average value were designated as swollen. 14 Plus (AB Technology, Pullman, WA). One or two embryos were deposited non-2 surgically into the uterine horn ipsilateral to the ovary with the CL. Pregnancy was 3 determined by palpation per rectum or ultrasound monitoring on d 70 after transfer. All 4 pregnancies were allowed to carry on to term. 5

Statistical Analysis 6
The data on nuclear swelling (nuclear area) were subjected to an arc sine 7 transformation. The transformed data were then analyzed by ANOVA (General Linear 8 Model, SPSS 11.0,Chicago,IL) [22] For the analysis of in vitro and in vivo development 9 of cloned embryos and the proportions of embryos that reached cleavage, developed to 10 the 8-cell stage, and to the blastocyst, as well as the conception rates on d 70 and calving, 11 were transformed by an arcsin transformation, and analyzed by a Student's t-test. A P 12 value of less than 0.05 is considered to be statistically significant. 13 Embryos 16 Cultured cumulus cells, presumably at G0/G1[18], were designated for use in our 17 nuclear transfer experiments. The nuclear areas of introduced donor cells, in IA, DA and 18 IAA groups, regardless of their activation regime, were found to increase progressively 19 and directly with the time interacted within a recipient oocyte (FIG.1). As a comparative 20 control, the average areas of cumulus nuclei, from donor cells prior to NT, were 21 measured as 61 ± 11 m 2 (n=29, FIG. 2A). There was no significant nuclear swelling in

Developmental Potential of Cloned Embryos Derived From Immediate or Delayed 19
Activation 20 The results of nuclear transfer (Table 1) indicated that a higher fusion rate was 21 observed in aged oocytes (Group IAA, fusion at 29 hpm) compared to young oocytes 22 (Groups IA and DA, fusion at 25 hpm). A significant improvement of cleavage (2-8 cells) rates, and subsequent development to blastocyst (50.0 % vs. 11.6 %) were achieved by 1 immediate activation (Group IA) when compared to delayed activation (Group DA). The 2 efficiency of NT in Group IA, as judged by the rate of blastocyst development based 3 upon the number of oocytes fused, reached as high as 59.1 %. Overall development of 4 cloned embryos in Group IAA was also significantly higher than that from Group DA 5 (40.7 % vs. 14.2 %) although the oocyte age of both treatments was same at the time of 6 parthenogenetic activation (29 hpm). Thus, immediate activation proved much more 7 effective than delayed activation; the latter resulted in a blastocyst development rate of 8 only 14.2% (P<0.05). 9 The cleavage between Groups IA and IAA was similar, and subsequent 10 preimplantational development to blastocysts, as well as overall blastocyst (BL)  11 percentage (BL/fused oocytes) was higher in Group IA compared to Group IAA, but it 12 was not significant (59.1% vs. 40.7%, P=0.089). 13 Based on the similar nuclear remodeling and progression results, and comparable 14 in vitro developmental potential between Groups IA and IAA, cloned embryos derived 15 from either IA or DA were subjected to embryo transfer (ET) to examine their viability to 16 term development. Table 2 indicates the pregnancy and calving data with either fresh NT 17 embryos or vitrified NT embryos, derived from fibroblasts and cumulus cells as nuclear 18 donors. In DA group, all NT embryos (cumulus and fibroblast origin) were freshly 19 transferred into recipients. In IA group, NT embryos derived from cumulus donors were 20 vitrified prior to ET. In contrast, a total of 113 fibroblast derived NT embryos in Group 21 IA were either freshly produced (n=49) or vitrified (n=64), subsequently transferred into 22 recipients (fresh ET, n=44, vitrified ET, n=32; total ET, n=76, Table 2). Established pregnancies, on d 70 of gestation, indicated no statistically significant difference in 1 established pregnancies between transfers of blastocysts from Group IA (cumulus, 22.2 2 %, n=18; fibroblast, 9.2 %, n=76; overall, 11.7 %), and Group DA (cumulus, 11.1 %, 3 n=18; fibroblast, 12.5 %, n=112; overall, 12.3 %) (P>0.05). There was no difference of 4 pregnancy on d 70 between fresh and vitrified embryos in Group IA when fibroblasts 5 were used as donor for NT (Table 2). There were 2 live calves born in each of Groups IA 6 (11.1 %) and DA (11.1%) in which cumulus cells were used as donor cells. When 7 fibroblast cells were used for NT, a high fetal loss was observed after d 70 of embryo 8 transfer in both Groups IA (4 abortions) and DA (9 abortions) ( Table 2). One stillborn 9 clone from each activation treatment was observed; 2 (2.6 %) and 4 (3.6 %) live clones 10 were delivered from Groups IA and DA, respectively. The overall term development to 11 live clones was 4.3 % for Group IA (4 live clones, n=94), and 4.6 % for Group DA (6 12 live clones, n=130), respectively. 13

14
Our nuclear transfer study in cattle clearly demonstrated that an occurrence of PCC, 15 presumably induced by MPF, is not essential for the effective reprogramming/remodeling 16 of a somatic nucleus introduced into the cytoplasm of an MII oocyte. The direct 17 exposure of a somatic genome to an MII oocyte rich in MPF, was proven to be sufficient 18 for the successful reprogramming of a differentiated somatic nucleus [3,7,15]. Nuclear 19 envelope breakdown (NEBD) and PCC have been shown to occur within a short duration 20 (usually 2-4 h) after a donor nucleus was transferred into an enucleated metaphase II 21 (non-activated) mammalian oocyte [2,5,8,12,23]. Nevertheless, the mechanism(s) 22 involved during the interaction of a donor nucleus with an oocyte's cytoplasmic 23 environment, containing high levels of MPF, and the extent of chromosomal remodeling 1 due to PCC, remain obscure and quite controversial. Several previous studies reported 2 that a prolonged exposure of a donor nucleus, particular a G0/G1 nucleus, to a non-3 activated oocyte, in order to induce PCC, was beneficial for nuclear reprogramming [14, 4 17, 23]. In mice, Wakayama et al. (1998) showed a high proportion of enucleated 5 oocytes developed to morulae/blastocysts when they were activated following a 6 prolonged exposure of the adult somatic nucleus to the oocyte's cytoplasm [8]. The 7 inclusion of a prolonged interval between nuclear injection and oocyte activation was 8 believed to be beneficial for both pre-and post-implantational development [24]. 9 Somatic pig clones could be produced by the combined approaches of simultaneous NT 10 fusion/activation and followed with serial nuclear exchange technology [25]; however, 11 pigs [26, 27] shared a similarity with mice [24] in that the induction of PCC was in 12 association with beneficial nuclear reprogramming. The proportion of reconstructed 13 porcine oocytes developing to the blastocyst stage was lower when activation was 14 immediate [14]. Wakayama et al. (1998) believed one of the key steps for successful 15 cloning of mice was to induce PCC and the subsequent pronuclear-like vesicle formation 16 in the injected nuclei [8]. It was believed that in other species, such as rats, the failure to 17 produce live clones was attributed to insufficient PCC induction [28]. 18 Nevertheless, our results show that PCC is not a necessary process for nuclear 19 reprogramming and subsequent embryo development in cattle. In both Groups IA and 20 IAA, when donor nuclei were exposed to the presumably MPF-rich cytoplasm for only a 21 short time prior to chemical activation, PCC was not observed. We did observe, however, 22 rapid nuclear swelling that might be capable of inducing nuclear de-differentiation into a 23 pronuclear-like stage in both groups. Our results are in agreement with those of Fulka et 1 al. (1996) [29], who reported that exposure of the donor nucleus to the non-activated 2 oocyte, even for a very short time, has beneficial effects on nuclear remodeling. Our 3 results indicate that the molecular remodeling of an introduced nucleus still occurs within 4 the cytoplast of an activated oocyte, along with the progressively increasing nuclear 5 swelling; however, inducing dramatic chromosomal structural reformation, such as PCC, 6 can, and likely should, be avoided [3]. Our present results demonstrate that direct nuclear 7 and cytoplasmic interactions are sufficient for reprogramming and subsequent embryo 8 development, regardless of the presence of PCC. 9 It has been known that the formation of PCC may lead to dramatic chromosomal 10 changes, possibly causing a range of DNA damage (fragmented chromatin, joined 11 chromatin, chromosomal breakage), or loss of chromosomes [29,30], especially when 12 the donor cell cycle is not compatible with that of the recipient oocyte [5]. Our results 13 showed that a prolonged exposure (up to 4 h) of a donor nucleus to the presumptively 14 high-MPF levels of the pre-activation oocyte was not beneficial for the preimplantational 15 blastocyst development of cloned embryos. The in vitro developmental potential was 16 significantly lower in Group DA (14.2%), compared to Groups IA (59.1%) and IAA 17 (40.7%) ( Table 1). The development of NT embryos, derived from cultured skin 18 fibroblasts, was also higher with immediate activation than in those with delayed 19 activation (Du and Yang, data not shown). The effects of immediate or delayed 20 activation in previous reports of bovine somatic cloning are controversial. Wells et al. 21 (1999) achieved a rate of 27.5% blastocysts when cultured adult mural granulosa cells 22 were exposed to a cytoplast for a prolonged period of 4-6 h prior to activation; however, a direct comparison between immediate activation and a 4-6 h incubation prior to 1 activation (delayed activation) was not performed in their study. The discrepancies in the 2 results from various studies might also be explained by differences in selecting somatic 3 donor cells (bovine ES-like cells vs. somatic cumulus cells) [31], age of oocyte recipients 4 [32], the experimental conditions [23] and protocols used [32][33][34]. Akagi et al. (2003) 5 found that the DA method improved in vitro development potential of NT embryos. In 6 Akagi et al.'s (2003) study, relative aged oocytes (24 hpm) were subjected to immediate 7 activation; in our experiment, young oocytes at 21 hpm were arranged for IA treatment. 8 In our IAA treatment, oocytes at a similar age (25 hpm) were used for NT and subsequent 9 activation. In this case, the age difference between immediate and delayed activation 10 groups was eliminated at the time of activation (IAA and DA, activation at 29 hpm). As 11 shown in our results, Group IAA demonstrated significantly improved preimplantational 12 development in comparison to that in Group DA. While the pattern of nuclear 13 progression and remodeling in Group IAA was similar to what occurred in Group IA with 14 relatively young oocytes, blastocyst development was higher in Group IA (P=0.089). We 15 believe that young oocytes had more competent capability to reprogram an oocytes if 16 better oocyte activation and pre-implantational culture were used [15,21]. On the other 17 hand, recently, more reports demonstrated that the proportion of embryos with normal 18 chromosomal ploidy decreased as the incubation time prior to activation was prolonged. 19 Decreased blastocyst development (0-8.6%) was reported when the exposure period was 20 longer than 3 h [17]. A plausible explanation for inferior in vitro development of NT 21 embryos in cattle may be abnormal chromatin structure and/or anomalous pronuclear 22 formation [17] resulting in numerical chromosome errors, such as polyploidy and mixoploidy [35]. Our embryo transfer results demonstrated that vitrified NT bovine 1 embryos derived from immediate activation had similar in vivo survivability as embryos 2 derived from a delayed activation treatment. This suggests that reprogramming events, in 3 the absence of PCC, did not have a detrimental effect on in vivo viability of cloned cattle 4 embryos. In fact, our immediate activation procedure had resulted in fully developed and 5 healthy newborns in the past [16,19,36]. In the present study, 4 and 6 live clones were 6 produced from either immediate or delayed activation, respectively, indicating the further 7 implantational development potential was equivalent in the resultant cloned blastocysts 8 regardless of the manner of activation. Nevertheless, the efficiency of generating cloned 9 blastocysts was significantly increased with immediate activation (40.7-59.1% vs. 14.2%). 10 Nuclear reprogramming is a complicated process, involving not only cellular 11 nuclear-cytoplast interaction [5], but also epigenetic modification and molecular 12 differentiation [13]. It is believed that the processes of nuclear swelling, remodeling of 13 the somatic nucleus, and chromosomal modification were required for successful 14 reprogramming [3]. NT oocytes in Group DA had significantly larger nuclear sizes than 15 those in Groups IA and IAA during 12-24 hpf. We assumed that during the process of 16 reformation of a pronuclear-like structure from the PCC chromatin phase, swelling 17 factors existing in cytoplasm are likely to more readily participate in nuclear 18 reconstruction. As a result, dramatically enlarged nucleus/nuclei could be formed in the 19 DA group. Because only 91.2% of the NT oocytes were observed to show PCC in Group 20 DA, we cannot also exclude the possibility that some of oocytes in Group DA actually 21 did not involve PCC-dependent nuclear reformation. A small proportion of oocytes in 22 Group DA might directly develop nuclear swelling similar to that which occurred in Group IA; however, this proportion was believed to be relatively minimal. The 1 molecular reprogramming factors, or at least the reprogramming initiation molecules, 2 certainly reside in the cytoplast of the matured oocyte, and their function may vanish 3 post-parthenogenetic activation [15]. MPF activity in an oocyte was reported to be at a 4 basal level 1 h post activation [3,21]. MPF or mitogen-activated protein kinase activity 5 was shown not to be a direct regulatory factor for reprogramming in cattle [3]. However, 6 our results cannot rule out the possibility that MPF is acting as an initiator of 7 reprogramming because a somatic nucleus was introduced into a cytoplast containing 8 high concentrations of MPF [21,37]. In the cases of our Groups IA and IAA, the nuclear 9 envelope was observed to be intact and the entire area of the nucleus had expanded 10 during activation. We hypothesize that the reprogramming factors can be incorporated 11 into the nucleus during nuclear swelling without disrupting the nuclear envelope 12 membrane; thus, reprogramming of the nucleus occurs without dramatic chromosomal 13 restructuring, such as PCC. 14 Cattle may represent a unique and excellent species as a model to extrapolate 15 about humans, while they appear to be distant from mice and pigs with respect to 16 immediate and delayed activation. At the very least, improving reprogramming in the 17 context of somatic cell NT requires erasure of cellular memory inherent in the donor cell 18 and re-establishment of patterns of gene expression and their regulation, such as 19 epigenetic methylation and acetylation, for competent embryogenesis and differentiation 20 [38]. One report demonstrated that DNA methylation supported intrinsic epigenetic 21 memory in mammalian cells [39]. They found that DNA methylation is not required for 22 the establishment of the maintenance of silent chromatin status; however, it conferred to the chromatin structure a long-term, intrinsic epigenetic memory that prevents gene 1 reactivation. Recently, a typical egg protein nucleoplasmin was reported to induce 2 massive chromatin decondensation that resulted in nuclear swelling, and to significantly 3 influence epigenetic modification, such as histone phosphorylation and acetylation [40]. 4 Although it is possible that molecules, such as nucleoplasmin, may serve to catalyze the 5 exchange of somatic and embryonic histones and de-repress gene expression, the long-6 term benefit has not been determined. A comprehensive study with molecular/cellular 7 mechanistic approaches in combination with a cattle cellular reconstruction system via 8 somatic cell nuclear transfer will help address interesting questions related to 9 reprogramming events. Among these questions are whether immediate activation 10 improves the process of demethylation and histone acetylation, or whether induced PCC 11 affects a delayed demethylation, alters exchange of DNA proteins, and suppresses de-12 repression of genes [41]. 13 According to our data, it was evident that the aster was associated with the 14 introduced nucleus, suggesting that the centrosome, or microtubule organization center 15 (MTOC), was introduced into the oocyte with the donor cell via membrane fusion. We 16 observed the duplication and splitting of the aster following NT in all Groups--IA, IAA 17 and DA. The aster initially formed at the nuclear poles as a fusiform structure during the 18 first mitotic phase. This phenomenon is similar to that described by Navara et al. [42]. 19 Due to the experimental design, we were able to observe that the formation and 20 distribution of asters was independent of parthenogenetic activation but was closely 21 associated with pronuclear-like structures (FIG. 2 and FIG. 3). We found it most 22 interesting that no aster formation was observed in reconstructed oocytes that were 23 induced to PCC by delaying activation. The mechanism(s) inherent in PCC that inhibits 1 or prevents aster formation is unknown and warrants further investigation. 2 In conclusion, our study has demonstrated that premature chromosome 3 condensation (PCC) is not an indispensable prerequisite for the competent 4 reprogramming of a differentiated somatic genome in cattle. The direct exposure of the 5 donor nucleus to the MII cytoplast, presumably containing high levels of MPF, for a 6 relatively short period of time, is sufficient to trigger a cascade of nuclear reprogramming 7 and developmental events. Higher efficiency of blastocyst development was obtained by 8 immediate activation. Similar in vivo developmental potentials of NT embryos derived 9 from substantially varied protocols (IA vs. DA), and the birth of live clones from both 10 treatments, suggest that cattle may represent a unique species with a greater plasticity 11 available for mechanical, biochemical, and physiological manipulation. 12