Genomic imprinting in ruminants: allele-specific gene expression in parthenogenetic sheep

Studies in the mouse have established that both parental genomes are essential for normal embryonic development. Parthenogenetic mouse embryos (which have two maternal genomes and no paternal genome), for example, are growth-retarded and die at early postimplantation stages. The distinct maternal and paternal contributions are mediated by genomic imprinting, an epigenetic mechanism by which the expression of certain genes is dependent on whether they are inherited from mother or father. Although comparative studies have established that many imprinted mouse (and rat) genes are allele-specifically expressed in humans as well (and vice versa), so far imprinting studies have not been performed in other mammalian species. When considering evolutionary theories of genomic imprinting, it would be important to know how widely it is conserved among placental mammals. We have investigated its conservation in a bovid ruminant, the domestic sheep, by comparing parthenogenetic and normal control embryos. Our study establishes that, like in the mouse, parthenogenetic development in sheep is associated with growth-retardation and does not proceed beyond early fetal stages. These developmental abnormalities are most likely caused by imprinted genes. We demonstrate that, indeed, like in mice and humans, the growth-related PEG1/MEST and Insulin-like Growth Factor 2 (IGF2) genes are expressed from the paternal chromosome in sheep. These observations suggest that genomic imprinting is conserved in a third, evolutionarily rather diverged group of placental mammals, the ruminants. Received: 13 May 1998 / Accepted: 16 July 1998     

Pronuclear visibility, development and transgene expression in IVM/IVF-derived porcine embryos

A total of 1550 zygotes was used to assess the timing of pronuclear visibility, embryo development following DNA microinjection, and transgene expression in IVM/IVF-generated porcine embryos. After centrifugation, pronuclei could be seen in 61.6% of zygotes. In 55.3% of these only 1 pronucleus was visible. Pronuclear visibility was highest at 20 h post-insemination. Zygotes were microinjected with 1 of 2 LacZ gene constructs driven by either the SV40 early promoter (pSVON) or the human cytoplasmic beta actin promoter (pbActinLacZ). Development and transgene expression were assessed after either 48 h or 7 d in culture. After 48 h, significantly more zygotes with a single visible pronucleus developed to the 8-cell stage than zygotes in which no pronucleus had been seen (43.0 vs 24.8%), while those with 2 pronuclei were intermediate (31.4%). After 7 d, no difference in development to the morula stage was observed between noninjected control embryos (25.5%) and embryos with 1 (21.0%) or 2 pronuclei (22.5%); however, the proportion of embryos reaching the morula stage in the nonpronuclear group was significantly reduced (9.1%). After 48 h in culture, transgene expression was significantly higher in embryos with 2 pronuclei at the time of injection than in those with 1 (36.4 vs 17.9%). After 7 d in culture, 41.5% of morulae derived from zygotes with 2 pronuclei and 29.97% of thsoe derived from zygotes with 1 pronucleus showed signs of transgene expression. At this stage, significantly more morulae expressed the pbActinLacZ than the pSVON transgene (43.8 vs 25.8%). More than 80% of putative transgenic morulae or blastocysts showed evidence of mosaicism. These results demonstrate that IVM/IVF porcine embryos are able to develop in culture and express a microinjected transgene.     

Effects of low temperatures on in vitro produced bovine zygotes

The objective of this research was to investigate the effects of cooling on the development of bovine zygotes. One-cell bovine embryos were maintained at 39°C (control), 20°C, 10°C, or 0°C for 5, 10, or 20 minutes, then cultured in vitro for 7 days and the proportion of embryos developing to the compact morula or blastocyst stage compared between different treatments. Duration of exposure time had no effect on development. Development rates to the compact morula or blastocyst stage were 3.9%, 11.4%, 17.4%, and 24.4% for zygotes maintained at 0°C, 10°C, 20°C, and 39°C, respectively, with differences in embryo yield between every treatment (P < 0.05). In a second experiment, bovine pronuclei (karyoplasts) and cytoplasts were cooled at 0°C or maintained at 39°C for 5 minutes. Pronuclear transplantation was then utilized to create 4 types of reconstructed embryos, those with: 1) non-cooled pronuclei and non-cooled cytoplasm, 2) non-cooled pronuclei and cooled cytoplasm, 3) cooled pronuclei and non-cooled cytoplasm, and 4) cooled pronuclei and cooled cytoplasm. The proportion of embryos developing to the blastocyst stage was highest when non-cooled pronuclei were transferred into non-cooled cytoplasm (18.9%), and similar to that of non-cooled, non-manipulated control zygotes (13.2%, P > 0.05). No embryos developed to the blastocyst stage when pronuclei (cooled or non-cooled) were transferred into cooled cytoplasm. However, zygotes with cooled pronuclei transferred into non-cooled cytoplasm yielded 4.5% blastocysts (P < 0.05). More embryos developed to the compact morula or blastocyst stage when non-cooled vs. cooled cytoplasm was utilized, regardless of whether the pronuclei were cooled (P < 0.05). These data demonstrate that pronuclei are more tolerant to low temperature exposure than is ovum cytoplasm. Mol. Reprod. Dev. 47:435–439, 1997. © 1997 Wiley-Liss, Inc.     

Genomic imprinting of the insulin-like growth factor 2 gene in sheep

A number of genes in the human and mouse genomes are subject to genomic imprinting, with selective inactivation of one allele of a gene in a parent-of-origin specific manner. One of the first imprinted genes identified was the Insulin-like Growth Factor 2 gene (IGF2), which promotes growth of the fetus and is expressed from only the paternal allele in most tissues in both the mouse and human. The aim of this study was to establish the imprinting status of IGF2 in sheep (Ovis aries). Sheep provide an interesting model to study imprinting, owing to differences in their placental development and the fact that they have been subject to strong artificial selection for various production traits. We report the identification of a length polymorphism in the transcribed 3′-untranslated region of the ovine IGF2 gene. This polymorphism was used to map IGF2 to sheep Chromosome (Chr) 21 and demonstrate that IGF2 is indeed imprinted in sheep, being expressed from the paternal allele. We also report that the developmental switch from imprinted IGF2 expression in the fetal liver to biallelic IGF2 expression in the adult liver, which occurs in the human but not mouse, also occurs in sheep. Differences in male- and female-specific recombination values reported around the IGF2 locus in the human were also observed around the ovine IGF2 locus. The techniques developed in this study will enable the imprinting status of IGF2 to be assessed in a variety of tissues and stages of development in normal sheep. Received: 3 October 1998 / Accepted 29 January 1999     

Development of sheep androgenetic embryos is boosted following transfer of male pronuclei into androgenetic hemizygotes

Androgenetic embryos are useful model for investigating the contribution of the paternal genome to embryonic development. Little work has been done with androgenetic embryo production in domestic animals. The aim of this study was the production of diploid androgenetic sheep embryos. In vitro matured sheep oocytes were enucleated and fertilized in vitro; parthenogenetic and normally fertilized embryos were also produced as a control. Fifteen hours after in vitro fertilization (IVF), presumptive zygotes were centrifuged and scored for the number of pronucleus. IVF, parthenogenetic, and androgenetic embryos (haploid, diploid, and triploid) were cultured in SOFaa medium with bovine serum albumin (BSA). The proportion of oocytes with polyspermic fertilization increased linearly with increasing sperm concentration. After IVF, there was no significant difference in early cleavage and morula formation rates between the groups, while there was a significant difference on blastocyst development between IVF, parthenogenetic, and androgenetic embryos, the last ones displaying poor developmental potential (IVF, parthenogenetic, and haploid, diploid, and triploid androgenetic embryos: 43%, 38%, 0%, 2%, and 2%, respectively). In order to boost androgenetic embryonic development, we produced diploid androgenetic embryos through pronuclear transfer. Single pronuclei were aspirated with a bevelled pipette from haploid or diploid embryos and transferred into the perivitelline space of other haploid embryos, and the zygotes were reconstructed by electrofusion. Fusion rates approached 100%. Pronuclear transfer significantly increased blastocyst development (IVF, parthenogenetic, androgenetic: Diploid into Haploid, and Haploid into Haploid: 42%, 42%, 19%, and 3%, respectively); intriguingly, the Haploid + Diploid group showed the highest development to blastocyst stage. The main findings of our study are: (1) sheep androgenetic embryos display poor developmental ability compared with IVF and parthenogenetic embryos; (2) diploid androgenetic embryos produced by pronuclear exchange developed in higher proportion to blastocyst stage, particularly in the Diploid-Haploid group. In conclusion, pronuclear transfer is an effective method to produce sheep androgenetic blastocysts.     

Sex specific X chromosome expression caused by genomic imprinting

The conflict theory of genomic imprinting predicts that imprinted genes are growth enhancing when paternally expressed and growth suppressing when maternally expressed. The expression pattern of autosomal imprinted genes generally fits these predictions. However, the conflict theory cannot easily account for the pattern of X-linked imprinting in humans and mice. This has led us to propose a novel hypothesis that X-linked imprinting has evolved to control sex specific gene expression in early embryos. The hypothesis links paternal X-imprinting (i.e. paternal copy silencing) to random X-inactivation and the retention of Y-linked copies, and links maternal X-imprinting to escape from random X-inactivation and the loss of Y-linked copies.The hypothesis offers a good explanation of the existing data on X-imprinted genes.     

Effect of microinjection time during postfertilization S-phase on bovine embryonic development

Microinjection into bovine zygotes was performed to evaluate the effects of the timing of injection during the phase of DNA replication on the subsequent in vitro development of embryos and expression of injected chicken β-actin promoter-lac Z gene construct. The period of DNA replication of bovine zygotes, determined by ³H-thymidine incorporation, begins between 12 hr and 13 hr postinsemination (hpi) of in vitro matured oocytes, reaches a maximum from 17 hpi to 19 hpi, and is complete by 21–22 hpi. Aphidicolin, an inhibitor of DNA polymerase alpha, was used to synchronize the pronuclei and the zygote population. Treatment with aphidicolin at 9–18 hpi arrested DNA replication without affecting formation of the pronuclei or embryo development. Cycloheximide, an inhibitor of protein synthesis, was used for nucleocytoplasmic resynchronization of the aphidicolin-treated zygotes. Microinjection was performed at 15 (early), 18 (mid), and 21 (late S phase) hpi. Embryonic development was affected following each of the three microinjection times. The development of zygotes injected at 18 hpi was significantly higher (P<0.01) after 5 days of culture than those injected at 15 hpi or 21 hpi. Expression of the marker gene was observed in the higher stage of development (>16 cells) only in the zygotes injected at 18 hpi. At the earlier stages of development, the proportions of embryos showing expression of the foreign gene were the same for all microinjection times. In aphidicolin- and cycloheximide-treated zygotes, expression of the marker gene followed the same curve as development, i.e., expression was low when injected early or late and higher (P<0.005) when injected in the middle of zygotic S phase. The ability of the embryos to survive microinjection and to express the marker gene as a function of hpi seems to be influenced mostly in the cytoplasm processing stage rather than the pronuclei processing stage. © 1995 Wiley-Liss, Inc.     

Imprinting defects in mouse embryos: stochastic errors or polymorphic phenotype?

Summary: Defects in expression of imprinted genes are believed to cause developmental abnormalities and play a role in carcinogenesis. To determine whether spontaneous imprinting defects may occur in mouse embryos, we studied the expression of two imprinted genes H19 and Igf2 in individual postimplantation 7.5 d.p.c. and 8.5 d.p.c. embryos. Biallelic expression of H19 was found in 1.6% of the embryos, whereas biallelic expression of Igf2 was found in 0.5% of the embryos. The loss of H19 imprinting (LOI) observed in a small fraction of early postimplantation embryos may be purely stochastic. Alternatively, since we never observed it in an inbred background, it may depend on genetic factors acting in trans. Either mechanism could explain the occurrence of polymorphic imprinting as well as the genesis of sporadic imprinting defects, including cancer. The frequency of LOI of H19 was higher than the incidence of sporadic imprinting disorders in humans (about 1 in 20,000). This contradiction may be explained by different incidence of imprinting errors in different imprinted regions of the genome, in different species, or by loss of the majority of nonmosaic embryos with imprinting defects before birth. genesis 31:11–16, 2001. © 2001 Wiley‐Liss, Inc.     

Differential expression of imprinted genes in normal and IUGR human placentas

Genomic imprinting refers to silencing of one parental allele in the zygotes of gametes depending upon the parent of origin. Loss of imprinting (LOI) is the gain of function from the silent allele that can have a maximum effect of doubling the gene dosage. LOI may play a significant role in the etiology of intrauterine growth restriction (IUGR). Using placental tissue from 10 normal and 7 IUGR pregnancies, we conducted a systematic survey of the expression of a panel of 74 “putatively” imprinted genes using quantitative RT-PCR. We found that 52/74 (~70%) of the genes were expressed in human placentas. Nine of the 52 (17%) expressed genes were significantly differentially expressed between normal and IUGR placentas; 5 were up-regulated (PHLDA2, ILK2, NNAT, CCDC86, PEG10) and 4 down-regulated (PLAGL1, DHCR24, ZNF331, CDKAL1). We also assessed LOI profile of 14 imprinted genes in 14 normal and 24 IUGR placentas using a functional and sensitive assay developed in our laboratory. Little LOI was observed in any placentas for 5 of the genes (PEG10, PHLDA2, MEG3, EPS15, CD44). With the 149 heterozygosities examined, 40 (26.8%) exhibited LOI > 3%. Some genes exhibited frequent LOI in placentas regardless of the disease status (IGF2, TP73, MEST, SLC22A18, PEG3), while others exhibited LOI only in IUGR placentas (PLAGL1, DLK1, H19, SNRPN). Importantly, there was no correlation between gene expression and LOI profile. Our study suggests that genomic imprinting may play a role in IUGR pathogenesis, but mechanisms other than LOI may contribute to dysregulation of imprinted genes.     

Effect of egg composition on the developmental capacity of androgenetic mouse embryos.

Analysis of the developmental capacities of androgenetic and gynogenetic mouse embryos (bearing two paternal or two maternal pronuclei, respectively) revealed a defect in blastocyst formation of androgenetic, but not gynogenetic, embryos that was a function of the maternal genotype. Androgenetic embryos constructed using fertilized eggs from C57BL/6 or (B6D2)F1 mice developed to the blastocyst stage at frequencies similar to those previously reported, whereas androgenetic embryos constructed with fertilized eggs from DBA/2 mice developed poorly, the majority failing to progress beyond the 16-cell stage and unable to form a blastocoel-like cavity, regardless of whether the male pronuclei were of C57BL6 or DBA/2 origin. This impaired development was observed even in androgenetic embryos constructed by transplanting two male pronuclei from fertilized DBA/2 eggs to enucleated C57BL/6 eggs, indicating that the defect cannot be explained as the lack of some essential component in the DBA/2 cytoplasm that might otherwise compensate for androgeny. Rather, the DBA/2 egg cytoplasm apparently modifies the incoming male pronuclei differently than does C57BL/6 egg cytoplasm. Several specific alterations in the protein synthesis pattern of DBA/2 androgenones were observed that reflect a defect in the regulatory mechanisms that normally modulate the synthesis of these proteins between the 8-cell and blastocyst stages. These results are consistent with a model in which cytoplasmic factors present in the egg direct a strain-dependent modification of paternal genome function in response to epigenetic modifications (genomic imprinting) established during gametogenesis and indicate that preimplantation development can be affected by these modifications at both the morphological and biochemical levels.     

Mouse Zygotes Respond to Severe Sperm DNA Damage by Delaying Paternal DNA Replication and Embryonic Development

Mouse zygotes do not activate apoptosis in response to DNA damage. We previously reported a unique form of inducible sperm DNA damage termed sperm chromatin fragmentation (SCF). SCF mirrors some aspects of somatic cell apoptosis in that the DNA degradation is mediated by reversible double strand breaks caused by topoisomerase 2B (TOP2B) followed by irreversible DNA degradation by a nuclease(s). Here, we created zygotes using spermatozoa induced to undergo SCF (SCF zygotes) and tested how they responded to moderate and severe paternal DNA damage during the first cell cycle. We found that the TUNEL assay was not sensitive enough to identify the breaks caused by SCF in zygotes in either case. However, paternal pronuclei in both groups stained positively for γH2AX, a marker for DNA damage, at 5 hrs after fertilization, just before DNA synthesis, while the maternal pronuclei were negative. We also found that both pronuclei in SCF zygotes with moderate DNA damage replicated normally, but paternal pronuclei in the SCF zygotes with severe DNA damage delayed the initiation of DNA replication by up to 12 hrs even though the maternal pronuclei had no discernable delay. Chromosomal analysis of both groups confirmed that the paternal DNA was degraded after S-phase while the maternal pronuclei formed normal chromosomes. The DNA replication delay caused a marked retardation in progression to the 2-cell stage, and a large portion of the embryos arrested at the G2/M border, suggesting that this is an important checkpoint in zygotic development. Those embryos that progressed through the G2/M border died at later stages and none developed to the blastocyst stage. Our data demonstrate that the zygote responds to sperm DNA damage through a non-apoptotic mechanism that acts by slowing paternal DNA replication and ultimately leads to arrest in embryonic development.     

DNA preparation method can influence outcome of transgenic animal experiments

Abstract

In our continuing quest to improve the efficiency of producing transgenic animals, we have compared the influence of two transgene purification techniques on the efficiency of creating transgenic sheep and mice. Three hundred eighty‐seven sheep zygotes and 2,737 mouse zygotes were microinjected with one of four transgenes. Transgenes were isolated from plasmid sequences either by agarose gel electrophoresis followed by gel extraction or by a single step sodium chloride gradient fractionation technique. Four transgenic sheep and 61 transgenic mice were produced. Both sheep and mice embryos responded similarly to transgene preparation methods. Overall, pregnancy rate was higher for recipients that received embryos injected with NaCl purified DNA (mean + SEM: 64 ± 7% vs. 38 ± 7%). Furthermore, offspring per zygote transferred (NaCl, 22 + 3% vs. Gel, 12 + 3%) and transgenics born per zygote transferred (NaCl, 3.9 ± 0.6% vs. Gel, 1.5 ± 0.6%) were higher when the NaCl purified DNA was used. However, the proportion of offspring born that were identified as transgenic did not differ between transgene purification methods. Transgenes responded differently to methods of preparation. One of the four genes yielded a significantly higher proportion of transgenics when the transgene was prepared by NaCl purification. These data suggest that on average the NaCl gradient purification technique results in a higher embryo survival rate to term for both sheep and mice, but the technique has no influence on rate of transgene integration.     

Rhodamine 123 as a cytoplasmic stain for mammalian zygotes

A mitochondrion-specific fluorescent dye, rhodamine 123, stains the cytoplasm of ova from mice, rabbits, sheep, cattle and pigs. Mouse zygotes stained with rhodamine 123 are often observed with areas of negative stain where the pronuclei are located. However, such areas of negative staining are not observed in zygotes from rabbits, sheep or cattle. In vitro viability of mouse zygotes stained with rhodamine 123 is high if observed with fluorescent microscopy for 5 min or less. Observation for more than 5 min results in a significant effect on embryo developmental potential in vitro.     

High-frequency generation of viable mice from engineered bi-maternal embryos

Mammalian development to adulthood typically requires both maternal and paternal genomes, because genomic imprinting places stringent limitations on mammalian development, strictly precluding parthenogenesis. Here we report the generation of bi-maternal embryos that develop at a high success rate equivalent to the rate obtained with in vitro fertilization of normal embryos. These bi-maternal mice developed into viable and fertile female adults. The bi-maternal embryos, distinct from parthenogenetic or gynogenetic conceptuses, were produced by the construction of oocytes from fully grown oocytes and nongrowing oocytes that contain double deletions in the H19 differentially methylated region (DMR) and the Dlk1-Dio3 intergenic germline-derived DMR. The results provide conclusive evidence that imprinted genes regulated by these two paternally methylated imprinting-control regions are the only paternal barrier that prevents the normal development of bi-maternal mouse fetuses to term.     

A genomic reservoir for Tnfrsf genes is developmentally regulated and imprinted in the mouse

Tumor necrosis factor receptor superfamily is composed of at least 26 members in the mouse, three of which exist as a cluster within the imprinted Kcnq1 domain on chromosome 7. Tnfrsf22, 23 and 26 contain typical cystein-rich domains and Tnfrsf22 and 23 can bind ligands but have no signaling capacity. Thus, they are assumed to be decoy receptors. The developmental expression profile of these genes is unknown and knowledge of their imprinting patterns is incomplete and controversial. We found that all three genes are expressed during mouse embryonic development, and that they have a strong maternal bias, indicating that they may be affected by the KvDMR, the Kcnq1 imprinting control region. We found expression of an antisense non-coding RNA, AK155734, in embryos and some neonatal tissues. This RNA overlaps the Tnfrsf22 and possibly the Tnfrsf23 coding regions and is also expressed with a maternal bias. We were interested in exploring the evolutionary origins of the three Tnfrsf genes, because they are absent in the orthologous human Kcnq1 domain. To determine whether the genes were deleted from humans or acquired in the rodent lineage, we performed phylogenetic analyses. Our data suggest that TNFRSF sequences were duplicated and/or degenerated or eliminated from the KCNQ1 region several times during the evolution of mammals. In humans, multiple mutations (point mutations and/or deletions) have accumulated on the ancestral TNFRSF, leaving a single short non-functional sequence.     

Ubiquitous expression and imprinting of Snrpn in the mouse

Snrpn is known to be abundantly expressed in rodent brain and heart, and in two separate studies with neonatal mouse brain it has been shown to be maternally imprinted, that is, the maternal allele is normally repressed. We now provide evidence on the expression profile and imprinting status of Snrpn throughout development. Using RT-PCR, we have established that Snrpn is further expressed at low levels in lung, liver, spleen, kidney, skeletal muscle, and gonads. Moreover, using mice with only maternal copies of Snrpn (maternal duplication for the chromosome region involved and parthenogenotes), we have shown that the gene is imprinted in all of these tissues and, generally, from the time the gene is first expressed at 7.5 days gestation. In contrast to the findings made with the imprinted genes, Igf2, Ins1, and Ins2, there is no evidence of tissue-specific imprinting in the embryo with Snrpn. Nor, as found with Igf2 and Igf2r, is there evidence of a window of biallelic expression between the germ line imprint and the time of gene repression. The absence of Snrpn expression in early embryos contrasts with the findings in ES cells.