|
|
||||||||
Regular Article |
a Department of Developmental Biology, National Institute of Agrobiological Sciences, Tsukuba, Ibaraki 305-8602, Japan
b Department of Animal Biotechnology, All-Russian Institute of Animal Genetics and Breeding, Saint Petersburg 189620, Russia
c Department of Biology, University of Pennsylvania, Philadelphia, Pennsylvania 19104-6018
| ABSTRACT |
|---|
|
|
|---|
developmental biology, early development, embryo, gene regulation
| INTRODUCTION |
|---|
|
|
|---|
Real-time imaging of transcriptional activity in live preimplantation mouse embryos that harbor a transgene, which encodes the secreted Vargula luciferase, has been reported [11]. In these experiments, the culture medium was assayed for luciferase activity. Thus, although multiple measurements can be made on a single embryo, this entails transferring the embryo to a new drop of culture medium. In addition, there are to our knowledge no studies in which transient expression of secreted Vargula luciferase has been assayed following microinjection of a plasmid-borne reporter gene.
Several properties of green fluorescent protein (GFP), which is derived from the Pacific Northwest jellyfish (Aequorea victoria), and which fluoresces when it is irradiated with UV light, make it an ideal in vivo marker of gene expression in a variety of organisms [1216]. For example, GFP requires no substrates or cofactors except oxygen [17] to fluoresce, it is relatively heat-stable and resistant to proteases [18], and it is essentially not toxic. Although toxicity due to overexpression has been reported in transgenic plants [19, 20], this is likely a general problem associated with protein overexpression rather than an attribute of GFP. In fact, no toxicity of GFP expression has been reported when GFP has been used for the noninvasive detection of gene expression in preimplantation embryos [2124].
The major strength of GFP as a tool for monitoring gene expression in living cells is that temporal dynamics of its expression can be observed and quantitatively measured. The basic strategy used to quantify GFP images is to quantify GFP fluorescence of a standard solution of purified GFP by microscopy methods and to compare this fluorescence with that obtained from images of GFP expressed in cells (reviewed in [25]).
We report here the development of a simple and fast method based on the quantification of GFP fluorescence that permits noninvasive multiple measurements of gene expression in the same preimplantation embryo. The method, which offers several advantages over existing methods to assay gene expression in preimplantation embryos, monitors in real-time the fluorescence derived from the expression of a plasmid-borne reporter gene that encodes for the expression of a short half-life form of enhanced GFP (enhanced GFP), d1EGFP.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Plasmid pd1EGFP-N1 (Living Colors; Clontech, Palo Alto, CA) consists of a destabilized enhanced green fluorescent protein (d1EGFP) gene fused to the cytomegalovirus (CMV) immediate early promoter. d1EGFP, which has a half-life of
1 h, is a fusion protein of enhanced GFP (EGFP) with the PEST domain of mouse ornithine decarboxylase that confers the short half-life [26]. Plasmid DNA was purified on Qiagen columns according to the manufacturer's instructions. The DNA was diluted to a concentration of 10 ng/µl.
Two picoliters of the DNA solution were microinjected into the male pronucleus of 1-cell embryos recovered from superovulated (C57BL6xCBA) F1 mice
23 h after hCG injection using standard procedures [27]. Before microinjection, cumulus cells were removed by a 2-min incubation with 100 IU/ml of hyaluronidase (Sigma, Tokyo, Japan) in M2 medium. The 1-cell embryos were then washed through four 50-µl drops of M2 medium. Two-cell embryos were obtained following overnight culture of 1-cell embryos, and one nucleus was microinjected
47 h after hCG injection. Before and after the microinjection procedure, embryos were cultured in M16 medium at 37°C in a humidified atmosphere of 5% CO2 in air. All micromanipulation procedures were performed with Narishige manipulators mounted on a Nikon Diaphot inverted microscope (400x) (Nikon, Tokyo, Japan) and a cell injector (CIJ-1, Shimadzu, Tokyo, Japan).
Sodium n-butyrate (Wako Ltd., Tokyo, Japan) was added to the culture medium to a final concentration of 1.5 mM to determine the effect of inhibition of histone deacetylase activity on the transcription.
Plasmid DNA Recovery and Semiquantitative Polymerase Chain Reaction
To estimate the amount of remaining plasmid DNA, embryos were removed at 0, 1, 5, 20, 40, 55, 70, 85, and 100 h after microinjection and subjected to semiquantitative polymerase chain reaction (PCR) analysis. Groups of 10 embryos were transferred to 10 µl of lysis buffer (1x ampliTag Gold PCR buffer containing 0.1 mg/ml proteinase K), and after a 5-min incubation at 55°C, the proteinase K was inactivated by a 10-min incubation at 85°C. Samples were then diluted to a final volume of 50 µl with ampliTag Gold PCR buffer containing 1.25 units of ampliTaq Gold polymerase (PE Applied Biosystems, manufactured by Roche, Branchburg, NJ), 200 µM dNTP mix, and 20 pmol EGFP-specific primers [23]. PCR was performed with a Perkin Elmer Cetus DNA Thermal Cycler 480. The PCR cycles included an initial denaturation step for 10 min at 95°C followed by 25 cycles at 95°C for 30 sec, 58°C for 30 sec, and 72°C for 40 sec, followed by 7 min at 72°C to ensure complete extension of the PCR product. The samples were maintained at 4°C until they were used for quantitative analysis of PCR product, which was measured following gel electrophoresis in 2% agarose-TAE gel and staining with ethidium bromide.
As a negative control, a group of 10 noninjected embryos were subjected to all procedures described above. For the positive control, a series of end-point dilutions of plasmid equal to the total amount of injected DNA was added to the noninjected embryos, which were then processed as described above. The detection limit is given in femtograms of plasmid DNA. The plasmid DNA concentration in each sample was estimated by comparison with the end-point dilution of the positive control.
Calibration of GFP Fluorescence
A 1 mg/ml solution of recombinant GFP (rGFP; Clontech) was concentrated under reduced pressure and then diluted in TE buffer (10 mM Tris-HCl pH 8.0, 10 mM EDTA) to produce rGFP concentrations that ranged from 3 mg/ml to 0.15 mg/ml, and the protein concentrations were verified using an Immuno Mini NJ-2300 microplate reader (NNIKK, Tokyo, Japan) equipped with a standard filter set (absorbance was measured at 490 nm). The intensity of fluorescent signal within the series of rGFP concentrations was verified using a Fluorescence Multi-Well Plate Reader CytoFluor 4000-2 (PerSeptive Biosystems, Inc., Framingham, MA) equipped with a Cyto Fluor standard filter set (excitation at 485/20 and emission at 530/25 nm).
Microneedles designed for pronucleus DNA microinjection and the injector (CIJ-1, Shimadzu) were used to inject 20 pl of the rGFP dilutions directly into the cytoplasm of the one-cell embryos to produce a GFP fluorescence standard curve. One-cell embryos injected with rGFP dilutions were kept in a drop of M2 medium when imaged for GFP fluorescence using the Nikon Diaphot 200 inverted fluorescent microscope equipped with a mercury arc lamp (100 W) and a DM505 fluorescein filter set (consisting of a 450490 nm excitation filter and a 520 nm emission filter). The difference in cytoplasmic green fluorescence from the different concentrations of injected rGFP was readily apparent by eye. Photoimages of fluorescing embryos were taken with a digital camera (Nikon FDX-35) attached to the fluorescent microscope and were analyzed on a Macintosh computer using the NIH Image program, which is available at the following Internet address: http://hrsb.info.nih.gov/nih-image/. Fluorescence signals in rGFP-injected embryos varied linearly with concentration of rGFP over the range 360 pg and were used to construct standard curves to quantify transient expression from the microinjected reporter gene. For each point on the standard curve the fluorescent signal was measured in at least 3 individual embryos injected with known amounts of protein; this verified the reproducibility of the correlation between values of the fluorescent signal and protein content.
Monitoring and Quantification of d1EGFP Expression
To perform noninvasive quantitative measurements of d1EGFP in single embryos, a series (35) of 5-µl drops of M2 medium each containing a single embryo injected with a known amount of rGFP that ranged from appropriate minimum and maximum values was set up beside the embryos that were to be analyzed (Fig. 1A). The intensity of the fluorescent signal in the embryos was determined after a brief UV light exposure; the intensity of UV irradiation was regulated by using ND2/ND4 reducing filters so that the autofluorescence from control noninjected embryos was not detectable. The signal intensity was then converted to the amount of protein expressed using the standard curve.
|
| RESULTS AND DISCUSSION |
|---|
|
|
|---|
The vast majority of the pronuclear-injected embryos showed a nonmosaic and uniform green fluorescent signal (Fig. 2). In some embryos, however, the signal was uneven among blastomeres and, considering that the bulk of transcription driven from the plasmid-borne reporter gene is from unintegrated copies, this could have been due to the unequal distribution of microinjected DNA among dividing blastomeres. If so, this occurred preferentially during the mid-preimplantation stages rather than during the first cleavage, because the unevenly fluorescing embryonic area (both stronger and weaker) was usually the minor area of the embryo.
|
To verify that expression of d1EGFP did not impair preimplantation development, we compared the developmental capacity of embryos expressing d1EGFP with the developmental capacity of embryos injected with buffer alone (sham). Analysis of the results of 4 experiments indicated that there was no significant difference in the developmental rate to the blastocyst stage between EGFP gene-injected embryos (74%; 142 out of 192 embryos that survived injection) and sham-injected embryos (79%; 38 out of 48 injected that survived injection). In previous studies, reduced developmental competence was observed among both the GFP-fluorescing and nonfluorescing embryos [22, 28, 29], suggesting that there is no direct link between reduced embryo viability and GFP/EGFP expression.
In order to monitor gene expression in the same embryo multiple times, it was necessary to establish conditions for UV irradiation that could detect the fluorescence but not inhibit development in vitro. Accordingly, we tested the influence of multiple UV exposures on embryo viability. Embryos were first exposed to UV irradiation at the two-cell stage followed by UV irradiation each day up to the blastocyst stage. Under our conditions of illumination, there was no apparent difference in the incidence of development to the blastocyst stage of these illuminated embryos to that of control embryos that were not exposed to UV light (Table 1). More important, multiple exposures to UV light did not affect the developmental capacity of embryos. Longer (120 sec), multiple exposures to UV light decreased the incidence of development to the blastocyst stage, and no development to the blastocyst stage was observed after multiple (510) and prolonged (120 sec) exposures to UV light at the 2-cell stage.
|
We next quantified dlEGFP expression in single embryos in which multiple measurements were made on the same embryo (Fig. 3). In these experiments the signal intensity present in the embryos was converted to the amount of protein, and the average dlEGFP content was calculated for groups of 2030 embryos (a total of
100 embryos in 4 experiments) for each data point during the time-course experiment. Embryos arrested in development during culture were eliminated from analysis.
|
A green fluorescent signal was first detected in embryos 40 h after microinjection of the male pronucleus (i.e., at the 4-cell stage). Others have reported expression of a plasmid-borne luciferase reporter gene within 2024 h following microinjection [1]. The delay that we observed may reflect a combination of a reduced sensitivity of detection of EGFP relative to luciferase and the small amount of plasmid DNA microinjected (0.02 pg), compared with the larger amounts (0.20.4 pg) injected in previous studies [1, 2]. Expression remained essentially constant up to 70 h following microinjection and then progressively declined until little if any expression was observed by 100 h postinjection (i.e., the blastocyst stage). This decrease likely reflected the degradation of the injected plasmid and not a decline in the transcriptional activity, because PCR analysis revealed a decrease in the amount of plasmid DNA (Fig. 4). We observed that within the first hour after microinjection a 30%50% decrease in the amount of plasmid DNA had occurred; as suggested previously [30], this decrease is likely due to the leakage of the injected DNA into the cytoplasm, where it is rapidly degraded. By the expanded blastocyst stage (100 h postinjection) only
5% of the plasmid DNA remains.
|
Injection of the plasmid into the nucleus of a two-cell blastomere resulted in the rapid appearance of detectable fluoresence that was initially higher than that observed following injection of the male pronucleus (Fig. 3). A rapid increase is also observed following injection of a 2-cell blastomere nucleus with the plasmid-borne luciferase reporter gene, and this level of expression is likewise greater than that observed following injection of 1-cell embryos [2]. Expression then increased and remained essentially constant for the next 40 h, after which it revealed a progressive decrease up to the blastocyst stage. Again, the decrease likely reflected the degradation of the injected plasmid, as determined by PCR analysis (Fig. 4), and not a decline in transcriptional activity.
A transcriptionally repressive state develops during the maternal-to-zygotic transition ([31] and references therein). This repression has been detected by the requirement for an enhancer for high levels of expression of a plasmid-borne luciferase reporter gene following injection into a 2-cell nucleus [3]. Moreover, inducing histone hyperacetylation by butyrate treatment relieves this requirement. It is interesting that expression of the plasmid following microinjection into the male pronucleus shows neither a requirement for an enhancer nor increased expression following butyrate treatment [32]. Accordingly, we examined the effect of inducing histone hyperacetylation with butyrate on expression of our reporter gene following injection of either the male pronucleus or two-cell blastomere nucleus (Fig. 3). Little stimulation of dlEGFP expression was observed in butyrate-treated 1-cell embryos, whereas a robust stimulation was found following injection of a 2-cell nucleus, when compared with embryos not treated with butyrate. Thus, monitoring dlEGFP fluorescence as a measure of reporter gene expression faithfully mimics the expression of a luciferase reporter gene and the effect of inducing histone hyperacetylation on reporter gene expression.
In summary, the method described in this report offers the investigator the opportunity to assay quantitatively in real-time the transcriptional activity of individual embryos under conditions that permit multiple measurements. The method should prove quite valuable in assessing the transcriptional activity in the preimplantation embryo (e.g., ascertaining the effect of different culture conditions on transcription, investigating the molecular mechanisms that underlie the maternal-to-zygotic transition, and analyzing transcription in species in which only a limited number of embryos are available [in nonhuman primates]).
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
1 This research was supported by the Japan Information Center of Science and Technology, the Organized Research Combination System Program, and by grant HD 22681 from the National Institutes of Health to R.M.S. ![]()
2 Correspondence: Serguei Yu. Medvedev, Department of Developmental Biology, National Institute of Agrobiological Sciences, Kannondai 2-1-2, Tsukuba, Ibaraki 305-8602, Japan. FAX: 81 298 38 8635; sergyurimedvedev{at}netscape.net ![]()
Accepted: January 30, 2002.
Received: December 4, 2001.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. G. Larman, C. M. Saunders, J. Carroll, F. A. Lai, and K. Swann Cell cycle-dependent Ca2+ oscillations in mouse embryos are regulated by nuclear targeting of PLC{zeta} J. Cell Sci., May 15, 2004; 117(12): 2513 - 2521. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |