|Home | About | Journals | Submit | Contact Us | Français|
The mouse preimplantation embryo development (Ped) gene product, Qa-2, influences the rate of preimplantation embryonic development and overall reproductive success. Here we investigated the expression pattern of two microRNAs, miR-125a and miR-125b, known to be involved in development in lower organisms, in preimplantation embryos from the two-cell, four-cell, eight-cell, morula, and blastocyst stages of development from the congenic B6.K1 (Ped negative) and B6.K2 (Ped positive) strains of mice.
B6.K1 and B6.K2 congenic mice differ only in the absence (B6.K1) or presence (B6.K2) of the genes encoding Qa-2 protein. We analyzed the expression of miR-125a and miR-125b in B6.K1 and B6.K2 preimplantation embryos by using real-time PCR.
We found no variability in miR-125b expression at any developmental stage in both strains. However, miR-125a expression increased during development in both strains and was ten times higher in Ped negative (B6.K1) embryos than in Ped positive (B6.K2) embryos by the blastocyst stage of development.
Our results show that the absence of the Ped gene profoundly affects the level of a miRNA (miR-125a) known to regulate early development. The implication is that miR-125a is likely involved in the regulation of timing of early development in mice.
MicroRNAs (miRNAs) are small non-coding RNAs approximately 22–25 nucleotides in length that function in regulating gene expression at the post-transcriptional level. miRNAs originate from genes that when transcribed form primary-miRNAs (pri-miRNAs). These pri-miRNAs are then cleaved by the protein Drosha in the nucleus into pre-miRNAs approximately 70 nucleotides in length that form imperfect stem-loop structures resulting from complementary base pairing in their sequences [1–3]. The pre-miRNAs are then exported out of the nucleus and into the cytoplasm by the protein Exportin 5 . Once in the cytoplasm the pre-miRNAs are cleaved again, this time by Dicer to form mature miRNAs . Dicer then recruits Argonaute protein family members to the mature miRNA and initiates the assembly of the RNA-induced silencing complex (RISC) [6–8]. The miRNA-RISC complex silences protein translation by base pairing with the 3′ untranslated region (3′UTR) of the target mRNA. For a detailed review of the biology of miRNAs see Bushati and Cohen  and Grimson et al. .
The first identified and probably the most well studied miRNAs are lin-4 and let-7. These miRNAs control developmental timing in the nematode worm C. elegans [11, 12]. The D. melanogaster homolog of lin-4 is miR-125b . In D. melanogaster, miR-125b, miR-100 and let-7 are upregulated in conjunction with metamorphosis .
Development in vertebrates has also been linked to miRNA expression. For example, in zebrafish, miR-430 family members have been found to regulate the transition from maternal genomic to zygotic genomic control of development as well as to rescue abnormalities of Dicer mutant embryos .
Similar to early development in C. elegans, D. melanogaster, and zebrafish, early development in mice is regulated by the expression of certain molecules at each stage of development. One gene in mice with a well documented role in early developmental timing is the preimplantation embryo development gene (Ped), which is associated with an increased rate of preimplantation embryo development in mouse preimplantation embryos that possess the Ped gene. Several reviews of the pleiotropic role of the Ped gene in development, reproduction, and overall adult health have recently been published [16–18]. In this paper we are concerned with the first described function of the Ped gene, the regulation of the timing of cleavage divisions in preimplantation mouse embryos.
In brief, it is now known that the Ped gene is located in the mouse major histocompatibility complex (MHC) [19–21]. The Ped gene product, Qa-2, a non-classical MHC class Ib protein, is encoded by four similar, tandem genes, Q6/Q7/Q8/Q9 [22–24]. The DNA sequences of the Q7 and Q9 genes and of the Q6 and Q8 genes are very similar so that these genes are referred to as the Q7/Q9 and Q6/Q8 gene pairs . Although the two gene pairs are slightly different from each other, the protein product of all four genes is called Qa-2 protein. In preimplantation embryos, only the Q7/Q9 gene pair is transcribed, although some mouse strains transcribe both genes (e.g. C57BL/6), some mouse strains, including the strain used in this study (B6.K2), transcribe only the Q9 gene [26, 27]. The availability of two congenic mouse strains that differ only in the presence (B6.K2) or absence (B6.K1) of the Q6–Q9 genes  has allowed extensive studies on Ped gene action without possible confounding effects of any other genes. When the Q9 gene is transcribed, Qa-2 protein is translated and expressed on the surface of preimplantation embryos. Expression of Qa-2 protein results in an embryo that divides faster (Ped fast phenotype) than an embryo in which there is a deletion of the Q9 gene and therefore lack of Qa-2 protein expression (Ped slow phenotype) [29, 30]. Embryonic rate of development is important to reproductive success because a fast rate of development is associated with a greater chance of survival to term.
In this paper we report the results of experiments designed to determine if miRNAs influence the regulation of Ped gene expression in preimplantation mouse embryos. To achieve this goal we submitted the sequence of the Q9 gene to the Sanger miRBase search engine (http://microrna.sanger.ac.uk) and determined that the Q9 gene has seven potential target sites (mmu-miR-709, -297, -297b, -668, -125a, -125b, -409, -215, and has-miR-595) for miRNA binding in its 3′UTR. Very interestingly miR-125 was listed as a potential regulatory miRNA with a target site in the 3′UTR of the Q9 gene. As mentioned above, miR-125 is the homolog of the C. elegans miRNA lin-4, which has been shown to be involved in the regulation of the timing of early C. elegans development. miR-125 exists in two forms, miR-125a and miR-125b, in most vertebrates, including zebrafish, mice, and humans.
In order to determine if the Ped gene (i.e. Q9) is influenced by miR-125 expression, we used quantitative real-time PCR to analyze miR-125a and miR-125b expression in preimplantation embryos from the two-cell, four-cell, eight-cell, morula and blastocyst stages of preimplantation development from the congenic B6.K1 (Ped negative) and B6.K2 (Ped positive) strains of mice.
The congenic B6.K1 and B6.K2 mouse strains were originally obtained from L. Flaherty (Wadsworth Center, Albany, NY) and subsequently bred in our laboratory. The only genetic difference between these two strains is in the Q region of the MHC, as shown in Table 1. The mice were housed in an AAALAC approved facility in a 14 h day/10 h night cycled room (lights on 0400–1800 EST) with controlled temperature and food and water ad libitum. All experiments followed the NIH guidelines.
Female B6.K1 and B6.K2 mice were superovulated with 5 IU eCG (Sigma Chemical Co., St. Louis, MO) at the ninth hour of the light cycle, followed 48 h later by 10 IU of hCG (Sigma Chemical Co.). Mice were mated, checked for vaginal plugs, and the plug-positive females were sacrificed by cervical dislocation. Embryos were collected 41, 53, 65, 77 and 96 h post-hCG corresponding to the two-cell, four-cell, eight-cell, morula and blastocyst stages of preimplantation embryo development, respectively. Embryos were collected in phosphate buffered saline (PBS) under 5% CO2, 5% O2, 90% N2, washed three times with PBS and then transferred in groups of 10 to 0.5 ml Eppendorf tubes containing 10 μl of lysis buffer (Stratagene, La Jolla, CA). The embryos were stored at −80°C until RNA isolation was performed. A total of 20 embryos was collected (two samples of 10 embryos each) for each stage of preimplantation development for each strain.
Mouse embryonic stem cells (ES cells), derived from C57BL/6 blastocysts, were purchased from Open Biosystems (Huntsville, AL) and used as a positive control. ES cells were maintained on a feeder layer of mitomycin c-treated primary embryonic fibroblasts (PMEFs) isolated from B6.K1 mice. To maintain an undifferentiated state, cells were grown in medium containing 80% DMEM, 15% fetal bovine serum, l-glutamine, 1 mM sodium pyruvate, 1% non-essential amino acids, penicillin/streptomycin, β-mercaptoethanol and 1,000 U/ml of leukemia inhibitory factor (LIF). The ES cells were grown in a 6 well dish until confluent. One confluent well was passaged into a T75 flask already containing PMEFS. The cells were grown for 3 days at which time they were trypsinized as follows.
Five milliliters of trypsin were added to the T75 flask once the growth medium had been removed. The flask was placed at 37°C for 2–3 min, and then 5 ml of growth medium were added to dilute the trypsin. Cells were transferred to a 15 ml tube and centrifuged at 930×g for 2 min. After centrifuging the cells, the medium was removed and the cells were resuspended in 4 ml of fresh medium. One microliter of the resuspended cells was used for cell counting. The cells were then divided into 1 ml aliquots in 1.5 ml tubes (1×106 cells/tube). Cells were stored at −80°C until RNA isolation was performed.
RNA was isolated from ES cells (1×106) using the Absolutely RNA Miniprep kit (Stratagene) following the manufacturer’s instructions. RNA was isolated from preimplantation embryos (10 embryos per sample) using the Absolutely RNA Nanoprep kit (Stratagene) following the manufacturer’s instructions. RNA from B6.K1 and B6.K2 preimplantation embryos was quantified using the Quant-iT RiboGreen RNA Assay Kit (Invitrogen, Carlsbad, California) following the manufacturer’s instructions.
Two microliters of ES cell total RNA and 1 ng of embryo total RNA served as the template for the 17 μl reverse transcription (RT) reaction. The reaction tube contained 1× PCR buffer II, 500 nM of each reverse primer (Table 2), 1.3 U RNase inhibitor, 16.75 U MMLV reverse transcriptase, 150 μM dNTPs, and 5 mM MgCl2 (GeneAmp RNA PCR Kit, Applied Biosystems, Foster City, CA). The reaction was carried out in a DNA thermal cycler 480 PCR instrument (PerkinElmer, Waltham, MA) using the following protocol: 16°C for 30 min, followed by 60 cycles of 20°C for 30 s, 42°C for 30 s, and 50°C for 1 s. A final step of 85°C for 5 min was added to inactivate the reverse transcriptase.
Five microliters of the RT reaction was used as template for the 20 μl Pre-PCR reaction. Each reaction tube contained 1.25 μM of each forward primer, 10 μM of the universal reverse primer (Table 2) and 1× SYBR Green PCR Master Mix (Applied Biosystems). The reaction was carried out in an ABI 7000 real-time PCR instrument (Applied Biosystems) under the following conditions: 95°C for 10 min, followed by 18 cycles of 95°C for 1 s and 65°C for 1 min.
Two microliters of undiluted Pre-PCR product was used as template in a 25 μl reaction. All samples were run in duplicate and each experiment was performed twice. Each reaction tube contained 1× SYBR Green PCR Master Mix (Applied Biosystems), 1 μM of each forward primer and, 1 μM of each reverse primer (Table 2). The real-time PCR protocol was as follows: 95°C for 10 min followed by 45 cycles of 95°C for 15 s and 60°C for 1 min.
The Delta Ct (threshold cycle) method of expression analysis was used following the instructions presented in Applied Biosystems User Bulletin #2 (http://www3.appliedbiosystems.com/cms/groups/mcb_support/documents/generaldocuments/cms_040980.pdf). As described later in the Results section, miR-125b expression was identical in all samples and was therefore used as an endogenous control to which miR-125a expression was normalized. miR-125a expression in the B6.K2 2-cell sample served as the calibrator to which all other samples were compared.
miR-125 is the homolog of C. elegans miRNA lin-4. It exists in two forms: miR-125a and miR-125b, which differ in that miR-125a has a diuridine insertion and a U to C change . (Fig. 1a). The predicted binding pattern of miR-125a and miR-125b with the 3′UTR of the Q9 gene is shown in Fig. 1b. There is one binding site for each of these miRNAs in the 3′UTR of the Q9 gene. Due to their small size, detection of microRNAs is difficult. Recently, Chen et al.  and Tang et al.  have published techniques in which microRNA quantification is possible using a stem-loop RT-PCR system. In brief, total RNA is used in a RT reaction in which specific reverse primers for the miRNAs of interest are used that have an extra sequence on the 5′ end that forms a loop structure by base pairing to itself. This type of primer is used because the extra loop structure improves thermal stability of the product . Performing RT this way (without the addition of random hexamers or Oligo d(T)16) generates a cDNA library of the microRNA(s) of interest, not of the total RNA in the sample. The RT product is then used as template for a Pre-PCR amplification step which utilizes the miRNA specific forward primers and a universal reverse primer that is specific for the 3′ end of the microRNA sequence generated in the RT reaction. The universal reverse primer specifically binds to the extra stem-loop sequence added during the RT reaction. Once this product is generated it is used as template for real-time PCR specific for each of the miRNAs of interest. Primers designed to amplify miR-125a and miR-125b were used in our experiments (Table 2 and Fig. 1c).
We first performed efficiency experiments on cDNA microRNA libraries created from ES cells covering four orders of magnitude (0.01 to 10 ng of total RNA input). miR-125a and miR-125b were detected in samples covering the four orders of magnitude tested: 0.01 to 10 ng of RNA input (data not shown). This analysis allowed us to determine our primer efficiencies (miR-125a=1.32, miR-125b=1.35) as well as validated our use of 1 ng of total RNA from B6.K1 and B6.K2 preimplantation embryos as template in the expression analysis experiments performed.
Real-time PCR detection of miR-125a and miR-125b was performed on two separate cDNA samples from each stage of preimplantation development from both the B6.K1 and B6.K2 strains of mice. The cDNA samples were generated using 1 ng of total RNA from a pool of 10 embryos and were run in duplicate during each of two replicates of the real-time PCR experiment. Therefore, for each stage of preimplantation development from each mouse strain there was a total of eight replicates. Average Ct’s (threshold cycles) were used for analysis of the levels of expression of the miRNAs.
Figure Figure22 is a representative example (single data point from each stage and strain from one experiment) of the amplification plots generated after amplification of miR-125a and miR-125b. As can be seen in Fig. 2a, there was no variation in amplification of miR-125b at any developmental stage in both B6.K1 and B6.K2 preimplantation embryos. However, amplification of miR-125a varied in a stage specific manner in both the B6.K2 (Fig. 2b) and B6.K1 (Fig. 2c) embryos. Since the cDNAs used in these experiments were not created using either random hexamers or Oligo d(T)16, amplification of a common endogenous control gene, such as GAPDH, for normalization was not possible. However, since miR-125b amplification did not fluctuate between developmental stages or between strains we used miR-125b expression to normalize miR-125a expression.
Figure Figure33 is of a representative agarose gel showing amplification of miR-125a and miR-125b in preimplantation embryos from the two-cell, four-cell, eight-cell, morula, and blastocyst stage from both B6.K1 and B6.K2 strains of mice (miR-125a = top panel, miR-125b = bottom panel). As can be seen, products are of the expected size (Table 2; ntc = no template control).
Figure Figure44 presents the normalized expression of miR-125a from embryos from each stage of preimplantation development (two-cell, four-cell, eight-cell, morula, and blastocyst) from both the B6.K1 and B6.K2 strains of mice. All samples were compared to the B6.K2 2-cell sample. The trend of expression is the same for both B6.K1 and B6.K2 embryos with an increase in expression through the blastocyst stage.
Figure Figure55 displays the ratio of expression of miR-125a in B6.K1 preimplantation embryos compared to B6.K2 embryos for each stage of preimplantation development. It is clear that miR-125a expression in embryos from B6.K1 mice is over ten times higher than expression in embryos from B6.K2 mice by the blastocyst stage of development.
The results presented in this paper demonstrate three important findings. First, miR-125b expression does not vary in a stage specific manner in mouse preimplantation embryos from the B6.K1 and B6.K2 strains of mice (Fig. 2a). Second, miR-125a expression does vary in a stage specific manner in mouse preimplantation embryos from both the B6.K1 (Ped negative) and B6.K2 (Ped positive) strains of mice (Fig. 2b and Fig. 2c). Third, miR-125a expression is over ten times higher by the blastocyst stage in preimplantation embryos from Ped negative compared to Ped positive mice (Fig. 5). These findings are the first example of expression differences of a particular miRNA being influenced by the presence or absence of a gene known to be involved in the regulation of preimplantation embryo development, namely the Ped gene. In addition, our results are also the first report of miRNA expression in morula and blastocyst stage mouse embryos.
A recent study has analyzed global miRNA expression in mouse oocytes and embryos up to the eight-cell stage of development . The data revealed that the total amount of miRNAs was downregulated by 60% between the one-cell and two-cell stages of preimplantation development followed by new miRNA expression beginning at the two-cell stage . They also found that elimination of the protein Dicer from oocytes yielded embryos missing virtually all miRNAs and created embryos unable to divide beyond the first cell division after fertilization. The authors concluded that maternal miRNAs are essential for proper zygotic development.
In the present paper we analyzed the expression of two specific miRNAs, miR-125a and miR-125b, in mouse preimplantation embryos. We chose these two miRNAs from the seven predicted miRNAs found in the 3’UTR of the Q9 gene because it had previously been reported that lin-4, the homolog of miR-125, is known to be involved in the timing of development in C. elegans  while miR-125b is known to be involved in morphogenesis of D. melanogaster .
In C. elegans, miRNA lin-4 controls the transition of larval stage L1 to L2 by negatively regulating the expression of the lin-14 and lin-28 proteins. If lin-4 is mutated, creating a loss of function phenotype, lin-14 and lin-28 levels persist and the worms display early larval cell fates later in larval development than normal . In addition, another miRNA, let-7, controls the transition from L4 to adulthood by negatively regulating lin-41 protein. If let-7 is mutated, lin-41 protein levels persist and the worms display abnormal larval stage fates in adulthood .
Developmental stage transitions in D. melanogaster are also characterized by miRNA expression pattern differences. Sempere et al.  have shown that miR-3, miR-4, miR-5, and miR-6 are downregulated in the transition from the embryonic to the larval stage. Downregulation of miR-10, miR-34 and miR-92, along with upregulation of miR-87, miR-100, miR-125b and let-7, define the transition from the larval to the pupal stage of development. Also, increased expression of miR-34 occurs at the onset of adulthood .
Similar to C. elegans and D. melanogaster, development of zebrafish has been found to be dependent on miRNA expression. Female fish that have had germ cells exhibiting a mutated from of Dicer injected into their embryos prior to fertilization produce embryos devoid of mature miRNAs and offspring that exhibit abnormal morphogenesis during gastrulation, somitogenesis as well as abnormal brain and heart development. Injection of miR-430 was found to reverse some of the abnormalities .
Our findings suggest that similar to lower organisms, miRNA expression likely plays a role in the regulation of preimplantation mouse embryo development. Interestingly, miR-125b expression differences were not found in our mouse embryos, which is a different result from the changes in miR-125b expression that are seen in D. melanogaster development. Rather we found that miR-125a expression increases during preimplantation mouse embryo development. Figure Figure1b1b shows that although the seed sequence (bases 2–7 of a miRNA) of both miR-125a and miR-125b match perfectly with the 3′UTR of the Q9 gene, miR-125a has three more predicted contact points with the 3′UTR of the Q9 gene than miR-125b. Based on our results of the expression pattern differences of these two miRNAs, it seems likely that these three extra contact points probably enhance miR-125a binding compared to miR-125b to the 3’UTR of the Q9 gene.
The challenge for future research will be to identify whether miR-125a affects the levels of expression of Qa-2 protein throughout the preimplantation period. The finding that the level of miR-125a increases with development is somewhat of an enigma since the level of Qa-2 protein expression also increases with development. The usual case with miRNA expression is that less target protein expression is associated with more miRNA. Therefore, the putative control of Qa-2 protein expression by miR-125a seems to be more complicated than the simple scenario of more miRNA-less protein. A search of the Sanger Institute’s miRBase shows that over 1,000 genes have potential binding sites for miR-125a. Therefore, we speculate that the regulation of the timing of early mouse embryo development is probably dependent on the interaction of a myriad of miRNAs and target genes.
Presence of the Q9 gene is necessary for Qa-2 protein expression in mouse preimplantation embryos, but Qa-2 protein has three other encoding genes (Q6, Q7, and Q8) involved in Qa-2 expression in other cells. Searching the Sanger miRNA database to see if miR-125a or miR-125b had predicted target sites for all four of the Qa-2 encoding genes revealed that only Q7 and Q9 but not Q6 and Q8 have target sites for miR-125a while Q7, Q8, and Q9, but not Q6, have predicted target sites for miR-125b (Table 3). This finding, along with the expression data presented here, suggests that miR-125a expression, but not miR-125b expression, is involved in development of mouse preimplantation embryos, possibly having a role in mediating the timing of early embryonic development as regulated by the Ped gene. Previously it was reported that the Q6/Q8 gene pair has a long-tandem repeat sequence in its 3′UTR not present in the Q7/Q9 gene pair (Table 3), which the authors suggested may be involved in the differential regulation of these genes . Our data show yet another difference between these gene pairs, namely presence of a miR-125a target site in the 3′UTR. Thus, expression pattern differences in embryos of the Q7/Q9 gene pair compared to the Q6/Q8 gene pair may be associated with regulation by miR-125a. However, it is still a mystery why some mouse strains express only the Q9 gene while others express both Q7 and Q9.
An intriguing result of our work is the finding that there is an inverse correlation between the presence of the Q6–Q9 genes and the level of miR-125a. It is not yet known whether the absence of these genes causes an increase in miR-125a levels or if the presence of these genes causes a decrease in miR-125a levels. There is precedence for the presence vs. the absence of the genes encoding the Ped gene phenotype affecting the expression of another molecule. Absence of Qa-2 expression has been shown to result in higher expression of platelet activating factor (PAF), a protein known to be beneficial to reproductive success . From the findings with PAF and the results reported in the present paper we can conclude that the presence or absence of the Qa-2 encoding genes can affect both protein and miRNA expression levels in preimplantation embryos.
The data presented here on the expression of miR-125a and miR-125b in mouse preimplantation embryos in the B6.K1/B6.K2 model system may be particularly relevant to the assisted reproductive technology (ART) clinic. The reason is that human embryos created by IVF or ICSI show a phenotype that recapitulates the Ped gene phenotype, namely increased pregnancy success from human embryos that develop at a fast rate compared to slower developing embryos [reviewed in 35, 36]. The human homolog of the Ped gene product, Qa-2, is HLA-G [37–40]. An exciting finding in the past few years, reported by five different research groups, is that the presence of soluble isoforms of HLA-G in the culture medium of embryos created after IVF or ICSI is correlated with an enhanced chance of pregnancy success [e.g. 41–45], although there are two groups that have not been able to repeat these results [reviewed in 46]. Regardless of whether or not soluble HLA-G ultimately turns out to be a clinically useful predictor of pregnancy outcome, expression of membrane-bound HLA-G, as well as membrane-bound Qa-2, is well documented to be associated with an enhanced rate of preimplantation development and overall reproductive success [reviewed in 47]. Currently, human embryo studies in the USA are not possible with federal funds. Therefore a similar study to that reported in this paper, the role of miRNAs in preimplantation human embryo development, is not possible at this time, further stressing the importance of our mouse model system.
This paper presents data on the expression of two particular miRNAs during preimplantation mouse embryo development. Since the discovery of miRNAs, over 5,300 miRNAs have been described, which are listed in the Sanger Institute’s miRBase. These miRNAs represent a whole new way of thinking about the regulation of gene expression. We are only at the very beginning of starting to understand the roles of these miRNAs in the regulation of mammalian development and reproduction.
In conclusion, we have shown that miR-125a, a miRNA, whose C. elegans homolog (lin-4) is temporally expressed in early developing C. elegans, and whose family member miR-125b is involved in D. melanogaster morphogenesis, is temporally expressed in early preimplantation mouse embryos with a dramatic increase in expression in embryos from mice with the absence of the Ped gene. Our results are the first example of expression differences of a particular miRNA, miR-125a, being linked to presence or absence of a gene known to be involved in preimplantation embryo development in mice.
We would like to thank Paula Lampton for technical assistance with the culture of the embryonic stem cells. We thank Michele Mammolenti and Robert Crooker for expert care of the mice. Supported by NIH grant HD39215, the Gordon Center for Subsurface Sensing and Imaging Systems (NSF EEC-9986821), and a NSF GK-12 pre-doctoral fellowship to M.B. (NSF-0338255)
Expression of miR-125a increases during preimplantation mouse embryo development and is 10 times higher in B6.K1 (Ped negative) mice than in B6.K2 (Ped positive) mice by the blastocyst stage of development.