Murine embryo development following cytoplasmic injection of linear and condensed DNA ()
1. INTRODUCTION
Page et al. [2], attempted to improve the rate of transgenesis by cytoplasmic injection by condensation of DNA with polylysine prior to injection into pronuclear stage mouse zygotes. Their efforts resulted in a transgenesis rate of 12.8% of the live animals tested. This certainly was an improvement over the 3.4% rate of transgenesis reported by Brinster et al. [1]. However, the new technique was demanding. Polylysine is a very large molecule and the large DNA aggregates frequently obstructed the microinjection needle pore (standard < 1 µm). Thus, a larger needle pore (>1 µm) became necessary to accommodate the transfer of the condensed DNA. The larger needle pore made the tip of the needle blunt by normal pore size comparison and microinjecting DNA with the larger needle reduced the survival of the embryos compared to microinjection of non-condensed DNA [2].
Complexing DNA with a smaller cation prior to cytoplasmic injection might greatly improve embryo survival because a smaller molecule would allow a smaller pore on the tip of injection needle. Monovalent cations do not condense DNA. Divalent cations require the addition of alcohol to facilitate greater cation-DNA interaction in an aqueous environment [3]. An added bonus to this process is that unlike condensation with poly-cations, aggregation is typically not observed when condensing with divalent cations [4]; fewer and smaller aggregates pass more easily through the injection needle into the cytoplasm. Greater embryo survival following cytoplasmic injection may result in an overall higher rate of transgenesis.
The objective of this study was to quantify the developmental effects of cytoplasmic injection of CMV-EGFP condensed with MgCl2.
2. MATERIALS & METHODS
2.1. DNA Preparation
The cytomegalovirus-enhanced green fluorescent protein (CMV-EGFP) construct was purchased already cloned into pIRES plasmid (Clonetech Laboratories, Palo Alto, CA, USA). The plasmid construct contained the CMVIE (immediate early) promoter (700 bp), followed by the multiple cloning site (MCS, 65 bp), the internal ribosomal entry site (IRES, 702 bp), the intervening sequence (IVS, 211 bp), and the EGFP complimentary DNA (cDNA, 1022 bp, including the polyadenylation sequence). Isolation and purification of the gene construct followed the general guidelines described in Chauhan et al. [5]. Isolation began with digestion of the plasmid overnight at 37.5˚C in 1× Universal Buffer (Stratagene, La Jolla, CA, USA) with endonuclease BamHI (Stratagene) to remove the IRES, IVS and a portion of the MCS. The restriction products were separated on a 0.8% agarose gel. The vector containing fragment was purified by Ultra Clean15 Kit, re-ligated with T4 DNA ligase (Stratagene) per manufacturer’s instructions, and used to transform E. coli XL-Blu competent cells (Stratagene) per manufacturer’s instructions which were then streaked on agar plates with Luria Broth and ampicillin (50 mg/L). Plasmid containing colonies were selected and cultivated in Terrific Broth (Gibco BRL, Carlsbad, CA, USA) with amplicillin (50 mg/L) over several days and several steps. The plasmid was then isolated and purified by column absorption (Qiagen Plasmid Maxi Kit; Qiagen Inc., Valencia, CA, USA). Next the plasmid was digested overnight at 37.5˚C in 1× Universal Buffer with NruI (Promega, Madison, WI, USA) and XhoI (Promega). The final construct which contained CMVIE promoter, followed by the truncated MCS (30 bp), and the intact EGFP cDNA. The total construct size was 1.752 kbp and had one sticky overhanging end left by the XhoI cut and one blunt end left by the NruI cut. The gene construct was isolated from a nondenaturing 1% agarose gel and purified by the UltraCleanTM15 DNA Purification Kit (Agarose Gels and Solutions, MoBio Laboratories Inc., Carlsbad, CA, USA). The purified construct was reconstituted in sterile deionized water. The DNA concentration was determined by photospectroscopy at 260 nm. The DNA was diluted to 5 and 625 µg/ml in intra-physiological saline (IPS) buffer comprised of 10 mM Tris HCl and 0.25 mM EDTA. Both diluted and undiluted DNA was stored at –20˚C until use.
2.2. DNA Condensation
The EGFP construct was condensed with MgCl2 according to a protocol developed by S. Butler (personal communication) to form rod shaped condensates [6] for cytoplasmic injection. The condensing medium contained 8 mM MgCl2, 10 mM Tris buffer, 40% deionized H2O, and 60% isopropyl alcohol. Condensation occurred during centrifugation at 15,000 × g for 45 min at room temperature. The DNA condensates were re-suspended overnight at 4˚C in injection medium which was intra-physiological saline (IPS) buffer containing 160 mM KCl, 20 mM MgCl2, and 5 mM NaCl at pH 7 (Butler, personal communication). Condensates were identified by dynamic light-scattering technology. The concentration was identified by photospectroscopy at 260 nm and then diluted with injection buffer to three concentrations for cytoplasmic injection: 100, 425, and 625 µg/ml. Condensates were then stored in injection buffer at 4˚C up to 7 d prior to use, or at –80˚C for later use.
2.3. Embryo Collection
Embryos were generated by superovulation, harvested, and cultured according to guidelines set forth by Hogan et al. [7]. Weaned female mice (strain CD-1) between 25 and 35 d of age, and weighing 19.5 to 22.5 g were administered 10 IU of equine chorionic gonadotropin (eCG, Diosynth, Chicago, IL, USA) intraperotineally (i.p.). These females were administered 5 IU human chorionic gonadotropin (hCG, Sigma Aldrich Chemical Company, St. Louis, MO, USA) i.p. 46 to 48 h later. The females were individually placed in cages containing a single male and 21 to 24 h later were examined for vaginal plugs, euthanized by cervical dislocation, and had their ovaries and oviducts removed for embryo harvesting. Embryos were collected at the cumulus stage and placed into 37.5˚C M2 medium (Specialty Media, Phillipsburg, NJ, USA) at pH of 7.4 [7]. The cumulus cells were dissociated from the embryos in M2 medium containing 1 mg hyaluronidase (320 IU/mg; Sigma Aldrich Chemical Company). All procedures were approved by the Animal Care & Use Committee.
2.4. Pronuclear Injection
Pronuclear microinjection was performed according to guidelines described by Canseco et al. [8]. Embryos were loaded into the injection chamber, a 100 mm Becton Dickinson Petri dish (Falcon, Franklin Lakes, NJ, USA) containing a narrow, centered droplet of M2 medium at 37.5˚C that was gently layered with mineral oil (Specialty Media) to reduce embryo exposure by media evaporation and temperature flux. Embryos were visualized and the male pronucleus was injected with 1 to 4 pl of DNA at 5 µg/ml. Following manipulation, the surviving embryos were cultured 4 d at 37.5˚C in 5% CO2 equilibrated CZB medium [9].
2.5. Cytoplasmic Injection
The injection chamber used for cytoplasmic injection was the same design as the one used for pronuclear injections. For cytoplasmic injection, the mouse embryos were loaded into the injection chamber, visualized, and injected in the cytoplasm, avoiding the pronuclear area. The volume of DNA injected was 10 to 15 pl [2] and the copy number injected ranged between 3750 and 5652 (Butler, personal communication). The murine embryos received one of six treatments: no injection, IPS injection buffer, 625 µg/ml linear CMV-EGFP in IPS buffer, or condensed CMV-EGFP in IPS with 160 mM KCl, 20 mM MgCl2, and 5 mM NaCl at 100, 425 or 625 µg/ml condensed CMV-EGFP. Following manipulation, surviving embryos were cultured 4 d at 37.5˚C in 5% CO2 equilibrated CZB medium [9].
2.6. In Vitro Embryo Development
Embryo development was assessed on d 4 and given a numerical score between one and four. Degenerated and lysed embryos were assigned the numerical score 1; one-cell and two-cell embryos were assigned the numerical score 2; three-cell to ten-cell embryos were assigned the numerical score 3; and embryos that progressed to the morula through blastocyst stages were assigned the numerical score 4. Data on embryos that received no injection were used to establish optimum development in the culture system. Embryos injected with IPS buffer were analyzed to establish the mechanical effects of the injection technique on embryo development.
2.7. Embryo Fluorescence
Embryos in the pronuclear injection study were evaluated for protein expression on d 4 as a percentage at each stage of development. Embryos in the cytoplasmic injection study were evaluated for protein expression as evidenced by fluorescence 4 d post cytoplasmic injection of CMV-EGFP. Fluorescing embryos were assigned a numerical score based on their stage of development, the same score design used to assess in vitro embryo development. Differences in embryo fluorescence between the treatment groups were then assessed as the percentage of embryos that showed protein activity in each treatment group and in the average development of the embryos that exhibited fluorescence across the treatment groups. To induce fluorescence, all embryos were subjected to several seconds of blue light at 488 nm, the excitation maximum for EGFP. Green fluorescent emission was examined in darkness at a wavelength of 508 nm (Nikon Filter set #EF-4 FITC HYQ, 480/(40) and 535/(50); Nikon Corporation, Tokyo, Japan).
2.8. Fluorescence Mosaicism
On d 4 embryos were also evaluated for mosaicism. Mosaic embryos did not exhibit fluorescence in every cell. Embryos with expression in every cell were identified as non-mosaic and were assigned the numerical score 0. Embryos with fluorescence in only some cells were assigned the numerical score 1. Mosaicism was evaluated by treatment and stage of development per treatment for both the pronuclear and cytoplasmic injection studies, and for stage of development across treatments for the cytoplasmic injection study. Data are expressed as least squares means.
2.9. Statistics
On d 4 embryos were evaluated and data collected for development, presence of fluorescence, and prevalence of fluorescence among blastomeres. Data were analyzed using one-way analysis of variance with SAS [10] or GraphPad Prism version 4.00 for Windows with Tukey's Multiple Comparison post test (GraphPad Prism, GraphPad Software, San Diego, CA, USA). Statistical models included treatment and residual effects. Bartlett’s statistic was used to test for equal variances.
3. RESULTS
3.1. Embryo Development in Vitro
Thirty-seven percent of pronuclear injected embryos lysed prior to culture. The process of pronuclear injection had a significant negative impact on embryo development (Table 1). Embryos that received no pronuclear injection had the highest (P < 0.05) mean development. Embryos injected into the pronucleus with either buffer or 5 µg/ml CMV-EGFP had similar (P > 0.05) development. Specific notation in numbers and percentages of embryos at each stage of development are in Table 2.
The in vitro evaluation of cytoplasmic injection as a means of producing transgenic mice included 3358 test zygotes. Three thousand and fifty three (90.9%) embryos were injected with either buffer or CMV-EGFP (condensed or linear) and the rest received the minimal manipulation and culture for 4 d. Of those injected, 62.8% lysed immediately (Table 2) and were discarded.
Control embryos totaled 305, of which 100% went into culture (Table 2). On d 4, the control group had the lowest percentage of degenerated and lysed embryos (Table 3) and the highest percentage to maximally develop to the morula to hatched blastocyst stages. The least squares mean development score of the d 4 control embryos was not different (P > 0.05) from the mean development score of the buffer treated embryos, which also had approximately 60% embryo development to the morula and hatched blastocyst stage.
Embryos in the group of zygotes injected in the cytoplasm with the transgene in linear form (625 µg/ml) had a low rate of degeneration and lysing (10.1%) and one of
Table 1. Distribution of embryo development 4 d following manipulation or pronuclear injection.
Table 2. Embryo lysing following cytoplasmic injections of CMV-EGFP.
Table 3. Distribution of embryo development 4 d following manipulation or cytoplasmic injection.
the highest percentages to develop to the morula to hatched blastocyst stage (32.0%, Table 3). The development score for this group was 2.8 ± 0.06, the highest of all the groups injected cytoplasmically. Similar mean embryo development was observed in the zygotes injected in the cytoplasm with 425 µg/ml condensed CMV-EGFP (2.8 ± 0.08). The embryos cytoplasmically injected with 625 µg/ml condensed CMV-EGFP had the highest rate of degeneration and lysing (23.5%) and the lowest percentage of embryos develop to the morula to hatched blastocyst stage (24.4%). The development score for this group was 2.3 ± 0.07, the lowest of all the groups injected. The groups of embryos injected with 100 and 425 µg/ml condensed CMV-EGFP had similar (P > 0.05) development scores (2.5 ± 0.07 and 2.8 ± 0.08, respectively) between the two-cell and three-cell to ten-cell stages.
3.2. Embryo Fluorescence
Forty percent of the pronuclear injected zygotes (79/196) exhibited fluorescence at 4 d in culture (Table 4). As the embryos in this treatment group developed, protein expression became more variable ending with 38%, 19%, and 25% non-mosaicism across the latter three stages of growth and statistically there was no difference between them (P > 0.05).
The zygotes injected in the cytoplasm with CMVEGFP, whether linear or condensed, exhibited the highest percentages of fluorescence in the one-cell embryos (56.5%). The lowest fluorescence was exhibited in morula and blastocysts (11.5%, Table 5). A calculation (total number of fluorescing embryos/total number of cultured embryos) revealed that only 26.9% of all zygotes injected in the cytoplasm with the DNA exhibited fluorescence, of those only 13.2% reached the highest stage of maturity.
Conversely, examination between the treatment groups and across all five stages of development, revealed the highest percentage of fluorescing embryos were in the zygotes injected in the cytoplasm with 625 µg/ml condensed CMV-EGFP (44.2%, Table 6), of which 16.7%, the highest among cytoplasmic injected zygotes, were morula and blastocysts (Table 7). The percentage of fluorescing embryos from cytoplasmic injection of 425 µg/ml condensed CMV-EGFP was similar (15.2%, Table 7) to that of zygotes injected in the cytoplasm with 625 µg/ml condensed CMV-EGFP; however, that number is skewed because the overall percentage of fluorescing embryos in that group was only 15.8% (Table 6).
Table 4. Embryo fluorescence and degree of mosaicism at five stages of development 4 d following pronuclear injection with CMV-EGFP (5 µg/ml).
Table 5. Fluorescence in embryos across treatments at all five stages of development 4 d following cytoplasmic injection with CMV-EGFP, both linear and condensed.
Table 6. Fluorescence at each stage of embryo development within treatment groups.
Zygotes injected into the cytoplasm with 625 µg/ml linear construct and 100 µg/ml condensed construct had the two the lowest total percentages of fluorescing embryos (24.3% and 23.6%, respectively, Table 6). Zygotes injected into the cytoplasm with 625 µg/ml linear construct also produced the lowest percentage of fluorescing morula and blastocysts (6.7%, Table 7). Similarly this treatment group had the lowest percentage of fluorescing degenerated and lysed embryos (11.7%), and the highest development score of fluorescing embryos across treatments (2.6 ± 0.10, Table 6) with 53.3% development to the three-cell to ten-cell stage (Table 7).
All three groups of zygotes injected in the cytoplasm with condensed CMV-EGFP had low mean development among fluorescing embryos which was lower (P < 0.05) than development among fluorescing embryos injected with linear DNA (Table 6). That means the biggest differences between the groups are observed in the highest overall percentage of fluorescing embryos, and in the highest percentage of maximally developed fluorescing embryos, both of which are seen in the group of zygotes injected into the cytoplasm with 625 µg/ml condensed CMV-EGFP.
Figure 1 is visual depiction of the differences and similarities between treatment groups of fluorescing embryos. Note the zygotes injected into the pronucleus with linear CMV-EGFP (5 µg/ml) had the lowest average development of fluorescing embryos.