MOLECULAR REPRODUCTION AND DEVELOPMENT 54:112–120 (1999)
Receptor-Mediated Transport of Foreign DNA Into Preimplantation Mammalian Embryos MARGARITA M. IVANOVA,1 ANDREY A. ROSENKRANZ,2,3 OLGA A. SMIRNOVA,2 VLADIMIR A. NIKITIN,2 ALEXANDER S. SOBOLEV,2,3 VLADIMIR LANDA,4 BORIS S. NARODITSKY,1* AND LEV K. ERNST2 1Laboratory of Genetic Engineering and Molecular Diagnostics of Microorganisms, Moscow, Russia 2Biophysical Laboratory, Institute of Agricultural Biotechnology, Moscow, Russia 3Department of Biophysics, Biological Faculty, Moscow State University, Moscow, Russia 4Institute of Molecular Genetics, Academy of Science of the Czech Republic, Prague, Czech Republic
ABSTRACT Mouse and rabbit preimplantation embryos with intact zona pellucida were incubated for 3 hr with DNA-carrying constructs containing insulin as an internalizable ligand: (insulin-polylysine)-DNA and (insulinpolylysine)-DNA-(streptavidin-polylysine)-(biotinylated adenovirus). Video-intensified microscopy demonstrated that the constructs penetrated the zona pellucida and accumulated in the blastomere perinuclear space. The percentage of blastocysts formed was about 70% after incubation of zygotes and two-cell embryos with the constructs. Foreign DNA was detected after 51 hr in 80% of rabbit embryos and after 96 hr in 73% of mouse embryos. Inclusion of various adenoviruses into the construct improved foreign DNA preservation in early embryos. Blot hybridization revealed genome-integrated foreign DNA in 12- and 15-day mouse embryos and in a newborn. Thus, the ligand-mediated mechanism can be employed for introducing foreign genetic material into early mammalian embryos; insulin provides for delivery inside the cell and to the nucleus, while adenoviruses ensure release from endosomes. Mol. Reprod. Dev. 54:112–120, 1999. r 1999 Wiley-Liss, Inc.
Key Words: adenovirus; embryos; gene transfer; insulin; receptor-mediated endocytosis INTRODUCTION Microinjection is currently a popular technique of introducing foreign genetic material into early embryos (Pursel and Rexroad, 1993; Echelard, 1996; Brenin et al., 1997). Notwithstanding the appreciable success gained with this approach, its high labor costs, the complexity of micromanipulations involving expensive equipment, high embryo lethality, and difficulty of locating the pronuclei in livestock embryos spur on attempts to design ways and means of obtaining transgenic animals without resorting to microinjection (Rottmann et al., 1985; van der Putten et al., 1985; BorisLawrie and Temin, 1993; Tsukui et al., 1995, 1996). Thus it has been proposed to use recombinant retroviruses as vehicles for the genetic material (van der Putten, 1985; Haskell and Boven, 1995), but in this case
r 1999 WILEY-LISS, INC.
the amount of the latter is naturally limited by the viral genome capacity, and such vectors are not easy to construct. In this context, it is desirable to transform embryos with a technique that does not require incorporation of the foreign DNA into the viral genome. Liposomes (Rottmann et al., 1985) and adenoviruses (Tsukui et al., 1995, 1996; Kubisch et al., 1997) have also been used to transfer genes into preimplantation embryos. Despite the increasing interest in noninjection techniques, none of these methods has yet become widespread. An actively developing approach to transgenosis of somatic mammalian cells makes use of receptormediated endocytosis for gene delivery (for review see Cotten, 1995; Guy et al., 1995). The component providing for permeation of the construct into the cell is an internalizable ligand. The presence of internalizable insulin receptors on early embryo cells (Heyner et al., 1989; Harvey et al., 1990, 1991) allows receptormediated gene transfer into preimplantation embryos, with insulin as the admission ligand in the DNAcarrying construct. In this work we tried to introduce foreign genetic material into preimplantation embryos in an insulin-containing construct that has been successfully used for transfection of cultured cells (Rosenkranz et al., 1990, 1992) and somatic cells in vivo (Sobolev et al., 1994, 1998).
Abbreviations used: Ad 5 neo, human adenovirus type 5, contains neomycintransferase gene; CELO, chicken embryo lethal orphan virus; EDS76, duck egg drop syndrome virus; EDTA, ethylenediaminetetraacetic acid; FITC, fluorescein isothiocyanate; HCMV, human cytomegalovirus promoter; Ins, insulin; pLys, polylysine; PCR, polymerase chain reaction; VIM, video-intensified microscopy. Grant sponsor: Russian State Subprogram ‘‘New Methods of Bioengineering; Genetic and Cell Engineering’’; Grant number: 150; Grant sponsor: Russian State Program ‘‘National Priorities in Health Care and Medicine, 08–Gene Therapy’’; Grant number: 08.01.01.04; Grant sponsor: Russian Foundation for Basic Research; Grant number 97-04-50181. *Correspondence to: Dr. Boris S. Naroditsky, Institute of Agricultural Biotechnology Russian, 127550, Timiryazevskaya St., 42, Moscow, Russia. E-mail:
[email protected] Received 13 July 1998; Accepted 14 May 1999.
RECEPTOR-MEDIATED GENE TRANSFER INTO EMBRYOS
113
MATERIALS AND METHODS Embryo Manipulations Inbred BALB/C mice and Chinchilla rabbits were used. Zygotes were collected from the oviducts in M2 medium (Sigma, St. Louis, MO) without removing cumulus cells. Two-cell embryos, morulae, and blastocysts were collected out from the oviduct and uterine horns with M2 medium. After transfection, the embryos were cultured in M16 medium at 37° under 5% CO2 (Hogan et al., 1986).
Preparation of Insulin-Polylysine-DNA Transfecting Construct Complex formation of the insulin-polylysine conjugate with plasmid DNA was carried out as described previously (Rosenkranz et al., 1992). Formation of virus-containing complexes was accomplished at pH 7.5 in 25 mM HEPES buffer with 150 mM NaCl and 0.25 mM EDTA by sequential addition of streptavidinpolylysine, plasmid, and insulin-polylysine to biotinylated viruses (Sobolev et al., 1998).
Synthesis and Purification of Conjugates Synthesis and purification of insulin-polylysine (InspLys) were carried out as described previously (Rosenkranz et al., 1992) with the use of a bifunctional cross-linking reagent N-succinimidyl 3-(2-pyridyldithio) propionate (Sigma). The conjugates were purified on a Sephacryl S-200 column (Pharmacia); the insulin/ polylysine molar ratio in different batches of the construct varied from 5 to 10. Streptavidin (Molecular Probes) was linked to 90-kDa poly-L-lysine with N-succinimidyl 3-(2-pyridyldithio)propionate as above; the conjugates were purified on a Sephacryl S-300 column (Sobolev et al., 1998); the polylysine/streptavidin molar ratio was 1 to 2.
Transfection In Vitro Embryos were incubated with the transfecting construct in 300 µl of M2 medium overlaid with mineral oil (Sigma) for 3 hr at 37°C and then washed with 300–500 µl M2 medium by pipetting in disposable tips (Sobolev et al., 1996). A portion of zygotes after transfection were transplanted into pseudopregnant foster mothers.
Plasmids The pGL2 control vector (Promega) contained the luciferase gene under the SV40 promoter; pVK28VP7 and pVK28VP7R (Akopian et al., 1992, Doronin, 1995) contained the VP7 gene for the porcine rotavirus outer capsid protein under the HCMV promoter; pCMV-LAluc (Sobolev et al., 1998) contained lactalbuminluciferase genes under the CMV promoter; pCMV-luc (Sobolev et al., 1998) contained the luciferase gene under the CMV promoter and was linearized with PvuI. Adenoviruses The duck egg drop syndrome (EDS-76) virus was grown as described previously (Rosenkranz et al., 1997). Chicken embryo lethal orphan virus (CELO) was grown in 9-day chicken embryos. After 3-day incubation, the virus-containing allantoic fluid was harvested (Laver et al., 1971, Li et al., 1984). Human adenovirus Ad5-neo (Van Doren et al., 1984) was grown in the 293 cell line. After incubation the infected cells were harvested, centrifuged, homogenized, and mixed with an equal volume of Freon 113 and then centrifuged again. The aqueous phase was used for virus purification (Winters and Russell, 1971). The virus-containing material was centrifuged through a discontinuous CsCl gradient. The virus bands were collected and centrifuged through a preformed linear gradient of CsCl. The virus bands were collected again and stored at 4°C. Virion concentration was determined spectrophotometrically (Chardonnet and Dales, 1970). Virions were biotinylated with biotinamidocaproate N-hydroxysuccinamide ester (Sigma) and dialyzed against 0.15 M NaCl, 25 mM HEPES, pH 7.5.
Transfection In Vivo Transfecting constructs were introduced into Balb/C females at the 3rd and 4th day of pregnancy (morula and blastocyst embryo stages). The construct was diluted 1:3 in M2 medium and introduced (50 µl) via a uterine probe for nonsurgical embryo transplanatation, or the construct was injected (10 µl) into the upper part of a horn of the uterus. At the first day of pregnancy, 5–10 µl of the construct was introduced surgically into the infundibulum of the oviduct. Use was made of Nembutal anesthesia (3–5 mg per mouse). DNA Isolation The washed embryos were transferred into tubes with 10 µl of deionized water under mineral oil. The micropipette capillary was filled with deionized water, and one embryo was withdrawn in a minimal volume of medium (1–2 µl). The embryo was disrupted by triple freezing-thawing and incubated for 1 hr at 37°C with 10 µg proteinase K (Serva). Then the enzyme was inactivated by heating (10 min at 100°C). To detect the transgene with PCR, fetuses and newborn mice were digested in buffer containing 50 mM Tris (pH 8,4), 100 mM EDTA, 0.5% sodium dodecyl sulfate, 200 µg of proteinase K per ml for 16 hr at 55°C, followed by extraction with phenol-chloroform-isoamyl alcohol and ethanol precipitation. DNA Analysis PCR of DNA isolated from one embryo was performed in a total volume of 25 µl of 67 mM Tris-HCL pH 8.8, 15 mM (NH4)2SO4, 2 mM MgCl2, 0.01% Tween-20 (Sigma), 170 µg/ml of bovine serum albumin (BSA), 200 µM dATP, dCTP, dTTP, dGTP, 25 µM each primer, and 1.25 units Taq polymerase (Fermentas, Lithuania) (Innis et al., 1990). The reaction mixure was added to microcentrifuge tubes with embryo genomic DNA. The PCR protocol was 94°C/5 min, 55°C/2 min, 72°C/2 min, then 40 cycles of 94°C/1 min, 55°C/1 min, 72°C/1 min. The pGL2 control vector (Promega) was analyzed using Promega primers (E1651 and E1661). The pCMV-luc
114
M.M. IVANOVA ET AL.
Fig. 1. Plasmid map pVK28VP7R and pVK28VP7. Empty are Ad5, filled are VP7, shaded are plasmid sequences. Numerals are units of the Ad5 genome map.
and pCMV-LA-luc plasmids encoding luciferase and human ␣-lactalbumin-luciferase fusion protein respectively (Sobolev et al., 1998) were analyzed with primers: direct 58-ATTGCTTTTGTGAGTATTTCTGTCTG, reverse 58-CCTTAGGTACCCAGTAGATCC. The primers for detection of the sequence of the porcine rotavirus surface antigen VP7 in pVK28VP7 and pVK28VP7R (Akopian et al., 1992, Fig. 1) were direct: 58-CTGGAGAGGAATTGGACATATCCG, reverse: 58-CCATCAACAACATCAACTATCGCC. To determine the specificity of the PCR product, the DNA from 2% agarose gel was transferred to Hybond N⫹ (Amersham) filter and hybridized (Maniatis et al., 1982) with [␥32P]-labeled probe 58TGCATCTCAATTAGTCAGC38. Then the autoradiographs were collated with the electrophoregrams. The positive PCR controls were pGL2 control vector and pVK28VP7 plasmids, 100 fg each. The negative controls were the reaction components without DNA, and genomic DNA of nontransfected embryos processed in parallel, as well as 10 µl of M2 medium from the last drop used to wash the embryo and the cultured M16 medium. Blot Hybridization Twenty micrograms of fetal DNA were cut with EcoRI, electrophorezed in 1% agarose gels, and transferred to Hybond N⫹ membranes (Amersham) by capillary blotting. The EcoRI fragment of pGL2control vector, pCMV-LA-luc was labeled by random priming using [32P]dCTP and used for blot hybridization at 68°C. The blots were washed under conditions for high stringency according to standard protocols (Amersham). DNA of newborns and pCMV-luc were cleaved with EcoRI and NcoI. Video-Intensified Microscopy Insulin-polylysine and streptavidin-polylysine conjugates were labeled with fluorescein isothiocyanate (FITC, Sigma) and purified as described earlier for insulin-polylysine (Rosenkranz et al., 1992) and DNAdelivering constructs were prepared. To ensure the absence of free pLys we used constructs with total lysine:nucleotide ratio 0.45 for the VIM experiments. Internalization of FITC-labeled constructs was assessed by video-intensified microscopy (VIM) using an
AT200 cooled CCD camera (Photometrics) and an Axioplan microscope (Zeiss), objective lens 40⫻ with numerical aperture 0.90. RESULTS Experiments In Vitro In order to investigate the interaction of DNAdelivering constructs with preimplantation embryos, FITC-labeled construct was prepared. The construct consisted of 5 nM pGL2 control vector plasmid and 63 nM Ins-pLys-FITC. During incubation of mouse embryos with FITC-labeled Ins-pLys-plasmid construct, the latter was accumulated in blastomeres (Fig. 2), and the process was attenuated by excess free insulin in the medium (Fig. 2B and D). VIM micrographs of preimplantation embryos carrying the construct did not reveal appreciable binding of the construct with zona pellucida (Fig. 2) at any stage of embryo development (from the zygote to eight blastomeres, data not shown). Nondeveloping ova obtained simultaneously with eightcell embryos could not take up the DNA-carrying construct. It is known that in the course of receptor-mediated endocytosis the endosome interior is quickly acidified, so the material trapped therein may then be digested by acid hydrolases (Smythe and Warren, 1991). An efficient means of preserving the internalized genetic material is to include an ‘‘endosomolytic’’ component into the carrier construct, the most potent of which is adenovirus (Cotten, 1995). VIM of embryos after incubation with a fluorescently labeled construct (Ins-pLys, 3.3 nM)-(pGL2 control vector plasmid, 0.5 nM)-(streptavidin-pLys-FITC, 3.0 nM)-(biotinylated adenovirus Ad5neo, 3.3 pM) demonstrated increased fluorescence in the perinuclear space relative to autofluorescence of control embryos, and no appreciable fluorescence of zona pellucida (Fig. 3). PCR was employed to assess the permeation of DNA into the cells of preimplantation embryos, its integrity and preservation during embryo development. A typical result of amplification after transformation with the pGL2control vector (5.3 kbase) construct is given in Fig. 4. No traces of DNA could be detected even in a tenfold volume of the medium taken with the embryo (Fig. 4, lane 8).
RECEPTOR-MEDIATED GENE TRANSFER INTO EMBRYOS
115
Fig. 3. VIM analysis of the virus-containing construct internalization into murine embryos: (a and b) embryo after 3 hr incubation with 0.5 nM (insulin-polylysine)-(pGL2 control vector plasmid)-(streptavidin-polylysine-FITC)-(Ad5-neo) construct; (c and d) control embryo (autofluorescence). Photographs a and c are transmitted light images: b and d are fluorescent images after 1 min integration.
Fig. 2. VIM analysis of the construct internalization into murine embryos: (a and b) embryo after 3 hr incubation in M2 medium at 37°C with 5 nM (insulin-polylysine-FITC)-(pGL2 control vector plasmid) construct; (c and d) same with excess free insulin 1 µM; (e and f ) control embryo (autofluorescence). Photographs a, c, and e are transmitted light images; b, d, and f are fluorescent images after 5 sec integration.
The construct without adenovirus was less efficient than the adenovirus-containing one implying that the latter is possibly less degradable because adenoviruses release DNA-delivering construct from endosomes. After 96-hr incubation foreign DNA was not longer detected in embryos, which had been transfected with Ins-pLys-DNA construct (Table 1, experiment 1). DNA proved to be transferred by adenovirus-containing constructs into both rabbit zygotes and two-cell embryos (Table 1), at a plasmid concentration down to 0.1 nM, i.e., less than 0.5 µg/ml. Similar transferability was observed with mouse zygotes and two-cell embryos (Table 2). We have earlier shown that the efficacy of transfection cultured cells with an (Ins-pLys)-DNA-(streptavidin-pLys)-(biotinylated adenovirus) construct depends on the lysine/ nucleotide ratio, the optimum being two (Sobolev et al.,
Fig. 4. PCR analysis of DNA from rabbit embryos after in vitro transformation at the two-cell stage: 1, buffer; 2–7, embryo DNA 51 hr after transformation with 5 nM construct; 8, embryo washing medium; 9–14, embryo DNA 51 hr after transformation with 0.5 nM construct; 16, pGL2control vector, 100 fg; 17, pGL2control vector, 10 fg.
1998). Such a construct provided successful DNA transfer in preimplantation embryos (Table 2, experiments 2 and 4). Avian adenoviruses CELO (Table 2) and EDS-76 (Sobolev et al., 1996) could also be used. Larger plasmids (pVK28VP7 and pVK28VP7R, about 10 kbase) could also be transferred into mouse embryo cells in vitro. Upon 48-hr incubation of zygotes with a construct comprising pVK28VP7R (1.5 nM), insulinpolylysine (16.2 nM), streptavidin-polylysine (15 nM), and CELO (5 pM), 8 embryos out of 9 contained the
116
M.M. IVANOVA ET AL. TABLE 1. Gene Introduction Into Rabbit Embryos In Vitro*
Constructa
Development stage at transformation
Time after transformation
No. of embryos
No. of embryos developed
No. of embryos with introduced gene
4 3
2 (50%) 0
6 6
6 (100%) 4 (66%)
Experiment 1 a b
1-cell 1-cell
96 96
8 8 Experiment 2
c d
2-cell 2-cell
51 51
9 10
*Embryos were incubated with the constructs (see Materials and Methods section) for 3 hr, washed (12 times), and grew for the indicated time in M16 medium. Isolated DNA was analyzed by PCR and blot hybridization as described in Materials and Methods section. aa, (pGL2control vector 0.5 nM)-(insulin-polylysine, 0.3 nM)-(streptavidin-polylysine, 0.7 nM)-(Ad5-neo, 0.17 pM); b, (pGL2control vector, 10 nM)-(insulin-polylysine, 47 nM); c, (pGL2 control vector, 5 nM)-(insulin-polylysine, 33 nM)-(streptavidin-polylysine, 30 nM)-(Ad5-neo, 33 pM); d, construct c dilution 1:10. TABLE 2. Gene Introduction Into Mouse Embryos In Vitro* Constructa
Development stage at transformation
Time after transformation (hr)
No. of embryos
No. of embryos developed
No. of embryos with introduced gene
2 3 12
2 3 12
2 (100%) 3 (100%) 9 (75%)
14 18
8 10
6 (75%) 7 (70%)
12 10
12 10
5 (40%) 1 (10%)
28
26
7 (26%)
Experiment 1 a a a
1-cell 1-cell 1-cell
3 24 48 Experiment 2
a b
1-cell 1-cell
96 96 Experiment 3
a a
2-cell 2-cell
3 72 Experiment 4
b
2-cell
72
*Embryos were incubated with the constructs (see Materials and Methods section) for 3 hr, washed (12 times), and grew for the indicated time in M16 medium. Isolated DNA was analyzed by PCR and blot hybridization as described in Materials and Methods section. aa, (pGL2 control vector, 5 nM)-(insulin-polylysine, 25 nM)-(streptavidin-polylysine, 35 nM)-(Ad5-neo, 33 pM); b, (pGL2control vector, 0.4 nM)-(insulin-polylysine, 6 nM)-(streptavidin-polylysine, 1.41 nM)-(CELO, 7.04 pM).
reporter gene; 72 hr after transfer of genetic material with the same construct into two-cell embryos, all three embryos tested gave a positive signal of the transferred DNA sequence. Transplantation of 35 embryos to pseudopregnant females yielded one fetus and two resorptions on the 11th day of development and two 15-day fetuses. Analysis of DNA by PCR with subsequent hybridization revealed positive signals of the VP7 gene in the 11-day fetus and one resorption in one 15-day fetus (Fig. 5). The transfer proved successful at all stages from zygote to morula. Thus 3 hr after incubation of murine morulae with the construct containing pGL2 control vector and Ad5-neo (construct and in Table 2), all six embryos tested were positive in the marker gene. Our preliminary data indicate that genetic material can also be transferred into sheep embryos after cryoconservation at the morula stage (Sobolev et al., 1996). Our experiments revealed no serious toxicity of the DNA-carrying constructs for mammalian preimplanta-
tion embryos in vitro. The survival with constructs averaged 70%, while in the negative control it was 80% from zygote to blastocyst and 83% from two blastomeres to blastocyst, both for mice and rabbits. Experiments In Vivo The foreign gene proved to be retained in the embryos at the postimplantation stages of development (Table 3). Blot hybridization was used to analyze the total DNA of 12-day fetuses (pCMV-La-luc in construct II into morulae), 15-day fetuses (pGL2 control vector in construct Ic into morulae and pCMV-La-luc in II into zygotes), and newborns (linearized pCMV-luc in construct III into morulae). The 15-day fetus obtained upon zygote transfection with construct II (Table 3) proved to carry foreign DNA in its genome. Blot hybridization (Fig. 6A) with 2-hr exposure revealed cleavage fragments of 1000, 1725, and 4400 bp; and with 24-hr exposure, an additional fragment of 10500 bp not produced upon cleavage of the plasmid. This proves
RECEPTOR-MEDIATED GENE TRANSFER INTO EMBRYOS foreign DNA integration into the embryo genome. Blot hybridization of the DNA of the 15-day fetus transfected with Ic construct revealed two fragments: 5100 bp (plasmid size) and 6000 bp with integrated foreign gene (Fig. 6B). In the newborn, blot hybridization revealed two fragments of 3000 and 600 bp, the smaller corresponding to pCMV-luc fragment and the larger most probably containing embryo DNA (Fig. 6C). For the first 12-day fetus, there were four fragments of 1000, 2068, 5632, and 8000 bp, the former three corresponding to the EcoRI fragments of pCMV-LA-luc and the latter containing the integrated gene (Fig. 6D, lane 4). The second 12-day fetus had the same three plasmid restriction fragments plus one of 4300 bp characteristic
of the given probe (Fig. 6D, lane 6). Blot hybridization of uterine tissue DNA (Fig. 6D, lane 5) demonstrated the 5632-, 2068-, and 1000-bp fragments indicative of episomal localization of the plasmid in somatic tissue. DISCUSSION The aim of the present study was to assess the feasibility of using receptor-mediated endocytosis to introduce foreign genetic material into preimplantation embryos. A prerequisite for this is inclusion of an internalizable ligand into the carrier construct. The presence of cell surface receptors in early embryos and the functioning of receptor-mediated endocytosis have thus far been demonstrated for a few ligands. Receptors for insulin and insulin-like growth factors (IGF-I and IGF-II) are expressed in preimplantation embryos (Schultz et al., 1993; Harvey et al., 1995), and exogenous insulin promotes cell proliferation and embryo morphogenesis (Harvey and Kaye, 1991, 1992; Pantaleon and Kaye, 1996). Therefore, we chose insulin as the ligand which can bind both to the cognate receptor and to the IGF ones. Again, DNA-carrying constructs with insulin have already been successfully used to transform cultured cells (Rosenkranz et al., 1990, 1992) and somatic cells in vivo (Sobolev et al., 1994, 1998). Although the modern techniques allow careful removal of zona pellucida, culturing the stripped early embryos, and introducing them into recipient animals, the possibility of transferring genetic material without such removal greatly facilitates the work and is extremely appealing. The permeability of zona pellucida for macromolecules and supramolecular complexes such
Fig. 5. Southern analysis of PCR-amplified genomic DNA for detection of the introduced VP7 gene. Transplantation of 35 embryos after transfection to pseudopregnant females yielded one fetus and two resorptions of the 11th day of development and two 15-day fetuses. Construct: pVK28VP7 (1.5 nM), insulin-polylysine (16.2 nM), streptavidin-polylysine (15 nM), CELO (5 pM). The 234-bp fragment is generated by the oligonucleotides in the PCR reactions. Lane 1, negative control; lane 2, DNA from 11-day fetus; lanes 3, 4, DNA from 11-day resorptions; lanes 5, 6, DNA from 15-day fetuses; lane 7, positive control containing pVK28VP7, 100 fg.
TABLE 3. Gene Delivery Into Mouse Embryos In Vivo* Constructa
Development stage at transformation
Time of analysis (days of embryogenesis)
No. of animals tested
No. of animals with introduced gene
8 8 3 6 6 8 10 5 2r
3 3 1 3 3 1 1 1 1
4 2 7 2 5r 39
1 2 2 0 1 1
A Ia
Blastocyst Blastocyst Blastocyst Blastocyst Blastocyst Morula Blastocyst Morula
5 8 11 15 8 15 8 15
II
Zygote Morula Morula Morula
15 12 11 15
III
Morula
Born
Ib Ic II
117
B
*Constructs introduced (A) via a uterine probe or (B) surgically into the oviduct of 1-day-pregnancy females and into a uterine horn of 3-day-pregnancy females. Total DNA of postimplantation embryos was analyzed by PCR with subsequent product hybridization and by blot hybridization (see Materials and Methods). r, resorptions. aIa, (pGL2control vector, 5 nM)-(insulin-polylysine, 25 nM)-(streptavidin-polylysine, 35 nM)-(Ad5-neo, 33 pM); IIb, construct Ia with adenovirus CELO; Ic, (pGL2control vector, 0.4 nM)-(insulin-polylysine, 6 nM)-(streptavidinpolylysine, 1.41 nM)-(CELO, 7.04 pM); II, (pCMV-La-luc, 3.6 nM)-(insulin-polylysin, 228 nM)-(streptavidinpolylysine, 40 nM)-(EDS76 33 pM); III, (pCMV-luc, 1.5 nM)-(insulin-polylysin, 230 nM)-(streptavidin-polylysine, 33.2 nM)-(EDS76, 33 pM).
118
M.M. IVANOVA ET AL.
Fig. 6. Southern analysis of genomic DNA from the fetuses and newborn. (A) Transfecting constructs II were introduced into females at day 1 of pregnancy. Genomic DNA of 15-day fetuses was digested with EcoRI, electrophoresed and blotted as described in Materials and Methods. Lanes 1 and 2, positive control pCMV-La-luc, 10 and 1 pg; lanes 3 and 6, DNA from the fetuses (exposure 2 and 24 hr, respectively); lane 4, DNA from the uterus; Lane 5, DNA from the fetus (negative control). The autoradiograph of blots was exposed for 2 and 24 days at ⫺70°C with an intensifying screen. (B) Transfecting constructs Ic were introduced into Balb/C females at the 3rd of pregnancy. Genomic DNA of 15-day fetuses was digested with EcoRI and HindIII, electrophoresed, and blotted as described in Materials and Methods. Lanes 1 and 2, positive control containing pGL2 control
vector, 10 and 1 pg; lane 3, DNA from fetuses; lane 4, fetus DNA (negative control); lane 5, DNA from the uterus. Autoradiograph of blots was exposed for 2 days at ⫺70°C with an intensifying screen. (C) Transfecting constructs III were introduced into females at the day 3 of pregnancy. Genomic DNA from the newborn mouse was digested with EcoRI and NcoI, electrophoresed, and blotted. Lane 1, DNA from the newborn; lane 2, DNA mouse (negative control); lanes 3, 4, positive control containing pCMV-luc, 0.1 pg and 1 ng. (D) Transfecting constructs II were introduced into females at the 3rd of pregnancy. Genomic DNA of 12-day fetuses was digested with EcoRI, electrophoresed and blotted. Lanes 1–3, positive control containing pCMV-Laluc, 100, 10, and 1 pg; lanes 4 and 6, DNA from the fetuses (1 and 2); lane 5, DNA from the uterus.
as viruses depends not only on their size but also on their interaction with the zona components. Thus the mouse zona pellucida is known to admit encephalomyocarditis virus (Gwatkin, 1967; Zusman et al., 1984); such a large virus as vesicular stomatitis virus cannot be removed from bovine embryos with trypsin and intense washing (Stringfellow et al., 1989). Enteroviruses and parvoviruses have been detected in the pores and sperm tract of pig embryo zona pellucida (Bolin et al., 1983). Zona pellucida is not a barrier for a number of macromolecules: in nature, maternal insulin is present in the reproductive tract and enters the embryo cells to influence their metabolism (Harvey and Kaye, 1990); DNA added to oocytes easily passes through the
zona pellucida (Chan et al., 1992). At the same time, zona pellucida is undebatably a barrier for many other viruses (Chen and Wrathall, 1989) and macromolecules (Legge, 1995). Its penetrability for viruses appears to vary in embryos of different species (Chen and Wrathall, 1989). Our VIM-aided direct experiments on penetration of murine embryo zona pellucida by the fluorescently labeled insulin-containing DNA-carrying construct show that the construct does not bind to zona pellucida but penetrates it to accumulate in the perinuclear space of the embryo cells at various stages from zygote to morula. For cultured hepatoma cells, we have observed no appreciable differences in internalization of insulin-
RECEPTOR-MEDIATED GENE TRANSFER INTO EMBRYOS polylysine-plasmid constructs in which either the DNA or polylysine was FITC-labeled (Murav’ev, Rosenkranz, and Sobolev, unpublished data). In embryo transfection, we used constructs with a lysine/nucleotide ratio less than 0.5; in this case all polylysine is bound to DNA. A serious problem in early embryo transformation is the damaging and/or toxic action of the components used in the process. As noted previously, insulin is a natural factor regulating embryo development. The effect of polylysine on early embryos has also been checked previously. Thus Page et al. (1995) have recently shown that a polylysine complex can be used in DNA microinjection into the cytoplasm of early embryos. Another polycation suitable for this purpose is histone H1, which can be injected into the zygote without disturbing embryo development (Lin and Clarke, 1996). It is known that inclusion of an adenovirus into the DNA-carrying construct enhances the transgene expression in cell culture (Curiel et al., 1991; Wagner et al., 1992; Fisher and Wilson, 1994), because in the course of receptor-mediated endocytosis the transferred DNA is entrapped in endosomes and then goes to lysosomes. The presence of components promoting the DNA release from endosomes, such as adeno- and rhinoviruses (Zauner et al., 1995), appreciably facilitates the delivery of the genetic material to the nucleus (Guy et al., 1995). To check the toxicity of adenoviruses for early embryos, we microinjected Ad5neo and Ad5 dl-312 into mouse and rabbit zygotes under the zona pellucida, into the cytoplasm, and in the pronucleus. Embryo development was within the norm, and survival was average for microinjection (30–40%) (Landa and Ivanova, unpublished). These data are in line with the recent reports about the applicability of adenoviral DNA as a vector in mammalian embryos (Tsukui et al., 1995, 1996; Kubisch et al., 1997). Our results indicate that adenoviruses of various groups, such as human and avian ones, can be used with equal efficiency and safety. The VIM fluorescent images testify to the import of the transfecting construct into early embryo cells. This process is competitively inhibited by the excess of free insulin in the incubation medium, indicating that the transport is mediated by insulin receptors. Thereupon the transferred DNA is retained, in both episomal and integrated forms, in the embryos of various stages, as evidenced by PCR analysis and blot hybridization. The irrefutable proof of the delivery of foreign genetic material by receptor-mediated transgenosis is the presence of integrated foreign DNA in the genomes of 12and 15-day fetuses and newborn animal. Our results demonstrate for the first time that receptor-mediated transfer of foreign genetic material can be applied not only to cultured or somatic cells but also to mammalian embryos of the first days of development. As already said, transgenic embryos harbor the foreign DNA in both integrated and episomal form. In the course of postimplantation development, the episomal foreign DNA appears to be eliminated, especially if
119
circular plasmids are used in the constructs; therefore the number of embryos and newborns carrying the foreign gene is reduced. Various methods of introducing genetic material into mammalian embryos give only a modest proportion of DNA integration (Canceco et al., 1994, Kubisch et al., 1995), leaving much to be optimized both in the techniques and in the material. The data obtained prove the basic feasibility of using receptor-mediated transport of foreign DNA into early mammalian embryos. Clearly, the genetic vector should be optimized to improve the integration efficiency. Apart from adenoviruses, promising agents for enhancing gene delivery via the ligand-mediated mechanism are amphipathic peptides (Plank et al., 1994) and gene-engineered protein complexes derived from adenoviral capsid (Fender et al., 1997). ACKNOWLEDGMENT We are indebted to Dr. S.E. Shaydulin for his invaluable help with rabbit embryos. REFERENCES Acopian TA, Lunin VA, Krugliak GG, Ruchadze VI, Bakhutashvili VI, Naroditsky BS, Tichonenko TI. 1992. Nucleotide sequence of the cDNA for porcine rotavirus VP7 gene (strain K). Virus Genes 6:393–396. Bolin SR, Turek JJ, Runnels LJ, Gustafson DP. 1983. Pseudorabies virus, porcine parvovirus, and porcine enterovirus interactions with the zona pellucida of the porcine embryo. Am J Vet Res 44:1036– 1039. Boris-Lawrie KA, Temin HM. 1993. Recent advances in retrovirus vector technology. Curr Opin Gen Dev 3:102–109. Brenin DR, Talamonti MS, Iannaccone PM. 1997. Transgenic technology: an overview of approaches useful in surgical research. Surg Oncol 6:99–110. Canceco RS, Sparks AE, Page RL, Russel CG, Johnson JL, Velander WH, Pearson RE, Drohan WN,Gwazdayuskas FC. 1994. Gene transfer efficiency during gestation and the influence of co-transfer of non-manipulation embryos on production of transgenic mice. Transgen Rec 3:20–25. Chan PJ, Su BC, Tredway DR, Seraj M, Seraj IM, King A. 1992. Uptake of exogenous human papilloma virus L1 DNA by oocytes and detection by the polymerase chain reaction. J Assist Reprod Genet 9:531–533. Chardonnet Y, Dales S. 1970. Early events in the interaction of adenoviruses with HeLa cells: I. Penetration of type 5 and intracellular release of the DNA genome. Virology 40:462–477. Chen SS, Wrathall AE. 1989. The importance of the zona pellucida for disease control in livestock by embryo transfer. Br Vet J 145:129– 140. Cotten M. 1995. Adenovirus-augment, receptor mediated gene delivery and some solution to the common toxity problems. Curr Top Microbiol Immunol 199(Pt 3):283–295. Curiel DT, Aqarval S, Wagner E, Cotten M. 1991. Adenovirus enhancement of transferrin-polylysine-mediated gene delivery. Proc Natl Acad Sci USA 88:8850–8854. Doronin KK. 1995. Expression of genes for porcine rotavirus VP7 and secreted alkaline phosphatase in the genomes of recombinant adenoviruses. PhD thesis. Moscow: Inst Agricultural Biotechnology. Echelard Y. 1996. Recombinant protein production in transgenic animals. Curr Opin Biotechnol 7:536–40. Erbacher P, Roche AC, Monsigny M, Midoux P. 1995. Glycosylated polylysine/DNA complexes: gene transfer efficiency in relation with the size and the sugar substitution level of glycosylated polylysines and with the plasmid size. Bioconjug Chem 6:401–410. Fender P, Ruigrock RWH, Gout E, Buffet S, Chroboczek J. 1997. Adenovirus dodecahedron, a new vector for human gene transfer. Nature Biotech 15:52–56.
120
M.M. IVANOVA ET AL.
Fisher KJ, Wilson JM. 1994. Biochemical and functional analysis of an adenovirus-based ligand complex for gene transfer. Biochem J 299:49–58. Guy J, Drabek D, Antoniou M. 1995. Delivery of DNA into mammalian cells by receptor-mediated endocytosis and gene therapy. Mol Biotechnol 3:237–248. Gwatkin RB. 1967. Passage of mengovirus through the zona pellucida of the mouse morula. J Reprod Fertil 13:577–578. Harvey MB, Kaye PL. 1990. Insulin increases the cell number of the inner cell mass and stimulates morphological development of mouse blastocysts in vitro. Development 110:963–967. Harvey MB, Kaye PL. 1991. Visualization of insulin receptors on mouse preembryos. Reprod Fertil Dev 3:9–15. Harvey MB, Leco KJ, Arcellana-Panlilo MY, Zhang X, Edvards DR, Schultz GA. 1995. Roles of growth factors during peri-implantation development. Hum Reprod 10:712–718. Haskell RE, Boven RA. 1995. Efficient production of transgenic cattle by retroviral infection of early embryos. Mol Reprod Dev 40:386– 390. Heyner S, Rao LV, Jarett L, Smith RM. 1989. Preimplantation mouse embryos internalize maternal insulin via receptor-mediated endocytosis: pattern of uptake and functional correlations. Dev Biol 134:48–58. Hogan B, Constantini F, Lacy E. 1986. Manipulation of the mouse embryo: a laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. Innis MA, Gelfand DH, Sninsky JJ, White TJ. 1990. PCR protocols: a guide to methods and applications. New York: Academic Press. 482 p. Kubisch HM, Larson MA, Lichen PA, Wilson JM, Roberts KM. 1997. Adenovirus-mediated gene transfer by perivitellin microinjection of mouse , rat, cow embryos. Biol Reprod 56:119–124. Laver WG, Younghusband HB, Wrigley NG. 1971. Purification and properties of chick embryo lethal orphan virus (an avian adenovirus). Virology 45:598–614 Legge, M. 1995. Oocyte and zygote zona pellucida permeability to macromolecules. J Exp Zool 271:145–150. Li P, Bellett AJ, Parish CR. 1984. The structural proteins of chick embryo lethal orphan virus (fowl adenovirus type 1). J Gen Virol 65:1803–15. Lin P, Clarke HJ. 1996. Somatic histone H1 microinjected into fertilized mouse eggs is transported into the pronuclei but does not disrupt subsequent preimplantation development. Mol Reprod Dev 44:185–192. Maniatis T, Fritsch EE, Sambrook J. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. Page RL, Butler SP, Subramanian A, Johnson JL, Velander WH. 1995. Transgenesis in mice by cytoplasmic injection of polylysine/DNA mixtures. Transgenic Res 4:353–360. Pantaleon M, Kaye PL. 1996. IGF-I and insulin regulate glucose transport in mouse blastocysts via IGF-I receptor. Mol Reprod Dev 44:71–76. Plank C, Oberhauser B, Mechtler K, Koch C, Wagner E. 1994. The influence of endosome-disruptive peptides on gene transfer using synthetic virus-like gene transfer systems. J Biol Chem 269:12918– 12924. Pursel VG, Rexroad CE. 1993. Recent progress in the transgenic modification of swine and sheep. Mol Reprod Dev 36:251–254. Rosenkranz AA, Yachmenev SV, Sobolev AS. 1990. The use of artifical constructions for selective transport of genetic material into human
cells by means of receptor-mediated endocytosis. Proc Acad Sci USSR 312:493–494 (in Russian). Rosenkranz AA, Yachmenev SV, Jans DA, Serebryakova NV, Murav’ev VI, Peters R, Sobolev AS. 1992. Receptor-mediated endocytosis and nuclear transport of a transfecting DNA construct. Exp Cell Res 199:323–329. Rosenkranz AA, Antonenko YN, Smirnova OA, Yurov GK, Naroditsky BS, Sobolev AS. 1997. Avian adenovirus induces ion channels in model bilayer lipid membranes. Biochem Biophys Res Commun 236:750–753. Rottmann OJ, Stratowa C, Hornstein M, Hughes J. 1985. Tissue specific expression of hepatitis B surface antigen in mice following liposome-mediated gene transfer into blastocysts. Zentralbl Veterinarmed [A] 32:676–82. Shultz GA, Hannel A, Arcellana-Panlilio M, Wang L, Goubau S, Watson A, Harvey M. 1993. Expression of IGF ligand and receptor genes during preimplantation mammalian delovepment. Mol Reprod Dev 35:414–420. Smythe E, Warren G. 1991. The mechanism of receptor mediated endocytosys. Eur J Biochem 202:689–699. Sobolev AS, Rosenkranz AA, Nikitin VA. 1994. Method for genetic transformation of animal mammary gland and device for introducing genetic material into the milk duct of animal mammary gland. Patent RU 2025487. Sobolev AS, Rosenkranz AA, Ivanova MM, Smirnova OA, Nikitin VA, Naroditsky BS, Ernst LK. 1996. Method of producing of the transgenic animals. Patent RU 2108714. Sobolev AS, Rosenkranz AA, Smirnova OA, Nikitin VA, Neugodova GL, Naroditsky BS, Shilov IA, Shatski IN, Ernst LK. 1998. Receptormediated transfection of murine and ovine mammary glands in vivo. J Biol Chem 273:7928–7933. Stringfellow DA, Lauerman LH, Thomson MS. 1989. Trypsin treatment of bovine ova after in vitro exposure to vesicular stomatitis virus. Am J Vet Res 50:90–992. Tsukui T, Mivake S, Azuma S, Ichise H, Saito I, Toyoda Y. 1995. Gene transfer and expression in mouse preimplantation embryos by recombinant adenovirus vector. Mol Reprod Dev 42:291–297. Tsukui T, Kanegae Y, Saito I, Toyoda Y. 1996. Transgenesis by adenovirus-mediated gene transfer into zona-free eggs. Mol Reprod Dev 42:291–297. Van der Putten H, Botteri FM, Miller AD, Rosenfeld MG, Fan H, Evans RM, Verma IM. 1985. Efficient insertion of genes into the mouse germ line via retroviral vectors. Proc Natl Acad Sci USA 82:148–52. Van Doren K, Hanahan D, Gluzman Y. 1984. Infection of eucaryotic cells by helper-independent recombinant adenoviruses: early region 1 is not obligatory for integration of viral DNA. J Virol 50:606–614. Wagner E, Zatloukal K, Cotten M, Kirlappos L, Mechtler K, Curiel DT, Birnstiel ML. 1992. Coupling of adenovirus to transferrin-polylysine/ DNA complexes greatly enhances receptor-mediated gene delivery and expression of transfected genes. Proc Natl Acad Sci USA 89:6099–6103. Winters WD, Russell WC. 1971. Studies on the assembly of adenovirus in vitro. J Gen Virol 10:1–194. Zauner W, Blaas D., Kuechler E, Wagner E. 1995. Rhinovirusmediated endosomal release of transfection complexes. J Virol 69:1085–1092. Zusman I, Engelhard D, Yaffe P, Ron A, Panet A, Ornoy A. 1984. Effects of interferon and encephalomyocarditis virus on in vitro development of preimplantation mouse embryos with and without the zona pellucida. Teratology 29:405–409.
