Transgenic Research

, Volume 19, Issue 5, pp 799–808

Transgenic rabbit production with simian immunodeficiency virus-derived lentiviral vector

  • L. Hiripi
  • D. Negre
  • F.-L. Cosset
  • K. Kvell
  • T. Czömpöly
  • M. Baranyi
  • E. Gócza
  • O. Hoffmann
  • B. Bender
  • Zs. Bősze
Original Paper

DOI: 10.1007/s11248-009-9356-y

Cite this article as:
Hiripi, L., Negre, D., Cosset, FL. et al. Transgenic Res (2010) 19: 799. doi:10.1007/s11248-009-9356-y

Abstract

Transgenic rabbit is the preferred disease model of atherosclerosis, lipoprotein metabolism and cardiovascular diseases since upon introducing genetic mutations of human genes, rabbit models reflect human physiological and pathological states more accurately than mouse models. Beyond that, transgenic rabbits are also used as bioreactors to produce pharmaceutical proteins in their milk. Since in the laboratory rabbit the conventional transgenesis has worked with the same low efficiency in the last twenty five years and truly pluripotent embryonic stem cells are not available to perform targeted mutagenesis, our aim was to adapt lentiviral transgenesis to this species. A simian immunodeficiency virus based replication defective lentiviral vector was used to create transgenic rabbit through perivitelline space injection of fertilized oocytes. The enhanced green fluorescent protein (GFP) gene was placed under the ubiquitous CAG promoter. Transgenic founder rabbits showed mosaic pattern of GFP expression. Transgene integration and expression was revealed in tissues derived from all three primary germ layers. Transgene expression was detected in the developing sperm cells and could get through the germ line without epigenetic silencing, albeit with very low frequency. Our data show for the first time, that lentiviral transgenesis could be a feasible and viable alternative method to create genetically modified laboratory rabbit.

Keywords

Laboratory rabbit Lentiviral transgenesis Simian immunodeficiency virus Mosaic expression Germ-line transmission 

Introduction

The rabbit is a standard laboratory animal in biomedical research and transgenic rabbits are used as animal models for a variety of human diseases both genetic and acquired. The rabbit (Oryctolagus cuniculus) is phylogenetically closer to primates than rodents (Graur et al. 1996) and is large enough to permit non-lethal monitoring of physiological changes. For these reasons, a number of research groups have chosen the transgenic rabbits as animal models for the study of lipoprotein metabolism, atherosclerosis, cardiovascular research and hypertrophic cardiomyopathy (Bosze and Houdebine 2006).

The first transgenic rabbits were obtained two decades ago by pronuclear microinjection (Hammer et al. 1985) and although improvements in the methodology have been reported since than (Besenfelder 1998) it is still an inefficient method: only 1% of injected eggs result transgenic founders. Creation of transgenic rabbits by somatic nuclear transfer would be a viable alternative, however, this method is still in infancy. Other alternative methods suffer from variability: recently dimethylsulfoxide-sperm-mediated gene transfer was adapted to produce human recombinant proteins in the milk of lactating does albeit at low levels (Shen et al. 2006).

Lois et al. (2002) and Pfeifer et al. (2002) opened the door for the use of replication defective lentiviral vectors for transgenic applications. The proof-of-concept study was performed by Lois et al. (2002) in mice and rat showing that the injection of VSV-G pseudotyped lentiviral vectors into the perivitelline space of fertilized oocytes could significantly increase production efficiency. Pfeifer et al. (2002) demonstrated that lentiviral vectors can be used to efficiently manipulate the zona-free embryos to produce founder transgenic mice. Since then human immunodeficiency virus (HIV-1)-derived lentiviral vectors were used to achieve reproducible high transgenesis rates in pigs (Hofmann et al. 2003), while transgenic chicken (McGrew et al. 2004) and transgenic pigs (Whitelaw et al. 2004) were created with an equine infectious anaemia (EIAV) based lentiviral vector. Transgenic cattle were successfully created by lentiviral infection of bovine oocytes but attempts with pre-implantation embryos failed (Hofmann et al. 2004). Rhesus monkey (Wolfgang et al. 2001) was the only species, in which the most commonly used HIV-based lentiviral vector injection did not result transgenic founders, however, recently it was successful in another non-human primate species, the common marmoset (Sasaki et al. 2009). In the last 5 years numerous transgenic laboratory and livestock animals were created by HIV-based lentiviral vectors and various tissue specific, cellular and viral promoters have been examined to direct transgene expression (Park 2007), but the creation of transgenic rabbits has not been reported so far. We aimed to use the lentiviral technology to increase the efficiency of transgenic rabbit production. In vitro data published earlier indicated that contrary to other mammalian cell lines, rabbit cells are poorly permissive to HIV-1 vectors, but could efficiently be infected with simian immunodeficiency virus (SIV) vectors (Hofmann et al. 1999). More recently, Cutino-Moguel and Fassati (2006) described a HIV-1 specific post-entry block in rabbit cells, which resulted aberrant trafficking of HIV-1 and was independent of the envelope construct used, suggesting that lentivirus vectors other than HIV-1 could be more appropriate to transduce rabbit cells. SIV-derived replication defective lentiviral vectors have now been generated in several laboratories. Characterization of these vectors showed that they are similar to HIV-derived vectors with respect to the insertion of transgenes in non-proliferating cells (Negre and Cosset 2002). SIV vectors perform better than HIV-1 vectors in simian cells (Negre et al. 2000) and were successfully used in the transplantation of genetically modified CD34+ cells to reconstitute the myeloid and lymphoid compartments (Derdouch et al. 2008). For creating transgenic animals we used SIV vectors for the first time and report here the establishment and germ-line transmission of the GFP marker gene in rabbits, created with SIV vector based lentiviral transgenesis.

Materials and methods

SIV-derived lentiviral vectors

A second generation packaging system and a monocistronic SIV transfer vector was utilized: the SIV-GAE-CAG-GFP-WPRE-SIN LTR (Negre and Cosset 2002). The CAG is a robust promoter, consisting of the CMV enhancer, the chicken beta-actin promoter and the rabbit beta globin intron. The method for generating SIV-based vectors has previously been described (Negre and Cosset 2002). Briefly, the virus particles pseudotyped with vesicular stomatitis virus G glycoprotein were produced by transient co-transfection of 293T vector producing cells using the calcium-phosphate method (Kvell et al. 2005). Their supernatant was concentrated 1000× fold by gradient-ultracentrifugation over a layer of 20% sucrose and titrated on Jurkat cells by measurement of GFP positive cells by FACS to establish transducing units per ml (TU/ml). The viral titre was found to be 1 × 106 TU/ml; the viral stock was re-suspended in DMEM and stored at −80°C.

Zygote collection and transduction

Collection of rabbit zygotes and transfer of injected embryos to recipient does was performed as described earlier (Bodrogi et al. 2006). Lentiviral vector injection was performed based on the modified perivitelline space injection method to produce transgenic mice from low titre lentiviral vector (Ritchie et al. 2007), injecting ~300–500 pl into the perivitelline space of each single cell embryo. Creation and handling of lentiviral transgenic rabbits was performed under biosafety level 2 precautions. All experiments were approved by the Animal Care and Ethics Committee of the Agricultural Biotechnology Center and complied with the Hungarian Code of Practice for the Care and Use of Animals for Scientific Purposes, including conditions for animal welfare and handling prior to slaughter.

Genotyping, identification of transgenic founders and their progeny, and determination of transgene copy number

Offspring derived from recipient does were screened for transgene integration by transgene specific PCR of genomic DNA purified from ear punch tissue or blood. GFP specific PCR was performed as published (Kvell et al. 2010).

Lentiviral transgene integration number was determined by Southern blot analysis of DNA from tail biopsy. 20 μg of DNA was digested with BamHI, separated on a 1% agarose gel, blotted to nylon membrane and probed with 282 bp-long PCR amplified GFP fragment using primers F: 5′-ctcgtgaccaccctgacctac-3′ and R: 5′-catgatatagacgttgtggctgtt3′.

Fluorescent imaging

GFP auto-fluorescence was detected using blue light illumination (GFP excitation frequency 455–495 nm with a barrier filter cut-off below 500 nm. For newborn and adult rabbits and wet tissues this was with a GFSP-5 headset (Biological Laboratory Equipment, Maintenance and Service Ltd., Budapest). Photomicrographs of the embryos were taken with Olympus BH2 research microscope, Zeiss Axio Imager microscope or using the MAA-03/B universal light source (Biological Laboratory Equipment, Maintenance and Service Ltd., Budapest) for Olympus SZH Stereo Zoom microscope.

GFP protein detection with western analysis

50–100 mg rabbit tissue samples were homogenized in eight volumes of ice cold PBS buffer. After centrifugation (12,000g for 1 min) the supernatants obtained from transgenic and non-transgenic animals were separated on 12% denaturing SDS–polyacrylamide gels. The proteins were transferred to PVDF membrane (Hybond-P, Amersham), blocked for 1 h with 5% nonfat dry milk. The blots were incubated for 1 h with a mouse anti-GFP monoclonal primary antibody (JL-8, Clontech, dilution: 1:10,000), washed, and then incubated for 45 min with a peroxidase conjugated sheep anti-mouse IgG secondary antibody (A5906, Sigma, dilution: 1:10,000). The blots were developed using the ECL-Advance chemiluminescence detection system (Amersham) in a dilution of 1:20 and Hyperfilm ECL autoradiography film (Amersham).

Results

Generation of transgenic rabbits

For viral transduction ~300–500 pl of the low titre (106 TU/ml) lentiviral vector was injected with repeated injections into the perivitelline space of fertilized rabbit zygotes, equal to one virus particle per zygote. After injection 291 embryos were transferred into eleven hormonally synchronized recipient females by endoscope. Altogether 87 rabbits were born from 9 does, among which 46 were stillborn. The pregnancy rate was high compared to the average 50% success rate of other pronuclear microinjection-based transgenic rabbit projects: 9 out of the 11 recipients delivered. Litter size varied from 3 to 15, which was notably higher, compared to the 2–4 average litter size in our earlier experiments (Bodrogi et al. 2006; Hiripi et al. 2003). Visualization of non-transferred embryos showed GFP expression at the blastocyst stage of lentivirus injected zygotes, which were cultured in vitro for 96 h (Fig. 1A).
Fig. 1

Detection of transgene expression in rabbit embryos and newborns. A SIV-CAG-eGFP lentivirus transduced rabbit embryos. GFP expression in blastocysts, cultured in vitro and derived from zygotes treated by perivitelline injection of 1 × 106 TU/ml lentiviruses. In vitro cultured rabbit blastocysts 96 h after lentivirus injection, under fluorescent (a, c) and conventional light (b, d). Arrowheads label GFP-positive blastocysts. B Different degree of GFP expression by direct epifluorescence in the skin of transgenic pups born from SIV-derived lentiviral vector injection, among its non-transgenic littermates not showing fluorescence

Transgenesis rate and expression of GFP transgene

Each rabbit born from the embryo transfers was analysed for the presence of the transgene by GFP specific PCR. Among the 87 offspring, 28 were found to be transgenic by PCR. GFP expression was evaluated only in the live-born transgenic founders (F0), first macroscopically by in vivo fluorescence imaging. Young, hairless founders showed individual patterns of mosaic green fluorescence as shown on Fig. 1B (a–d), suggesting that transgene integration occurred only at later stages of embryonic development. Eighteen from the 28 PCR positive live-born transgenic animals expressed GFP, indicating that the chromosomal integration site influenced the GFP expression as expected. Some of the adult transgenic animals showed green fluorescence in the areas not covered by hair: ears, nose and eyes (Fig. 4a), while others were not fluorescent macroscopically (data not shown). Southern analysis showed one transgene integration site (Supplementary Fig. 1), which was expected with this low viral titre. The transgenic animals appeared healthy, growth and reproduction rates did not differ from the transgenic rabbits and their non-transgenic littermates.

Variations in macroscopic GFP expression macroscopically and among the organs of the founders

Organs were excised for further analysis from one newborn founder following anaesthesia. In this founder the eyes and muscles showed very robust fluorescence, while the kidneys, heart and lung showed uneven GFP expression (Fig. 2). Age-matched non-transgenic littermate did not show fluorescence. The specificity of the transgene product was confirmed by western blot using a mouse monoclonal anti-GFP antibody. Derivatives of the three primary germ-layers were obtained from adult founder animals. Representative results are shown on Fig. 3. Analysis of the internal organs including liver, kidney (mesoderm), spleen, heart, cerebellum (neuro-ectodermal), lung, pancreas (endoderm) and skin (ectoderm) showed, that GFP expression-derived by a SIV-based lentivirus was achieved in all the organs examined, but not in all founder animals. It is important to note the harmony of our data: GFP-specific PCR of genomic DNA samples was negative from the organs of all founders in which RT–PCR or western blot analysis of GFP mRNA or protein expression was also found to be negative (RT–PCR data are not shown). Organ specific expression showed marked variance among individual founders in accordance with the mosaic GFP expression patterns observed in the animals. The quantity of GFP protein showed differences among different organs of the same animal and among the same organs (liver and lung) originating from different animals (Fig. 3).
Fig. 2

Profile of GFP expression in various tissues of a transgenic 8SIV3/1, Tsg) and a non-transgenic (Non-tsg). Rabbit born from SIV-CAG-eGFP lentiviral injection. All photos were taken under the same intensity of 489 nm excitation light (fluorescence) and visible light (white light)

Fig. 3

Western blot of tissue samples from organs of four different SIV-CAG-eGFP lentiviral transgenic founder rabbits. Variable level of GFP protein between the different animals and within the same organs. Note: the GFP protein in the pancreas of SIV9JKK, in the cerebellum of SIV6JT and SIV9BT, in the kidney and skin of SIV7JKK and SIV9JKK. GFP was visualized using a mouse anti-GFP monoclonal primary antibody (JL-8, Clontech, dilution: 1:10,000)

Germ-line transmission of the SIV-CAG-eGFP transgene

An important question is whether SIV-based lentiviral vector transgenesis results in germ-line transgenic animals. Semen samples of seven transgenic founder bucks were examined for the presence of the transgene by specific PCR and three of them were identified as germ cell transgenic. Sperm samples were obtained from three bucks and wild type females were fertilized by artificial insemination. One transgenic founder doe was inseminated with wild-type sperm and delivered 37 offsprings in five consecutive pregnancies. Table 1 summarises the number of progenies, the results of in vivo fluorescence examinations and the transgene specific PCR, performed from the ear biopsy of each progeny. Out of the total 232 offspring, the descendants of four mosaic GFP expressing founders, only one showed GFP expression in vivo. The wild type doe which carried this pregnancy was sacrificed at 13.5 days of pregnancy and photomicrographs were taken from the embryos. This embryo showed uniform green fluorescence (Fig. 4b2) and its yolk sac and placenta (Fig. 4b3 and 4c3) also expressed GFP. Non-transgenic embryos from the same pregnancy did not show fluorescence (Fig. 4c1, 4c2). Only two out of more than 200 live-born progenies were found to be transgenic by PCR and Southern analysis, but in those animals GFP expression was not detected, which could be the result of epigenetic regulation during germ-line transmission, as described earlier in a large animal model (Hofmann et al. 2006). Histological analysis of spermatogenesis in a transgenic founder male testis clearly showed that only a small fraction of spermatids express GFP in the seminiferous tubules, along with smooth muscle and Leydig cells (Fig. 5), which lines up with low transgenic offspring numbers observed (Table 1).
Table 1

Germ-line transmission rate in SIV-CAG-eGFP transgenic founders

Founders

In vivo GFP expressing/total born offspring

Transgenica/total born offspring

SIV3 BT male

1/70

1/70

SIV7JT male

0/85

1/85

SIV 9 JBK male

0/23

1/23

SIV10 JBK female

0/37

0/37

aBased on transgene-specific PCR and Southern analysis

Fig. 4

Detection of GFP expression in the SIV3BT founder rabbit and its progenies a GFP is strongly expressed in the ears, eyes and nails of the transgenic buck, b1b3, c3 GFP expressing transgenic embryo, c1c2 non transgenic embryo. 13.5 days old embryos in bright field (b1, c1) and fluorescent light (b2, c2), placenta (b3) and yolk sac (c3) of the transgenic embryo

Fig. 5

Spermatogenesis in the testis of the adult SIV3BT founder rabbit (saggital section of seminiferous tubules). GFP expressing spermatids, Leydig cells and smooth muscle cells were identified in the testis. a hematoxylin-eosin staining (a spermatogonia, b spermatids, c spermatids in maturation phase, d sperm, e smooth muscle cells, f Leydig cells). b Merge of GFP and DAPI staining (b GFP-expressing spermatids, e GFP-expressing smooth muscle cells, f GFP-expressing Leydig cells). c DAPI nuclear stain. d GFP expression with 488 nm excitation light. (Scale bar: 50μm)

Discussion

The present study examined the feasibility of using a SIV-based lentiviral construct to create transgenic rabbits. The first lentiviral gene-transfer vectors were derived from HIV-1 (Naldini et al. 1996). However, it is known that lentiviruses have a restricted species tropism. HIV-1 is potently restricted in rabbit cells at a post-entry stage, although infection by the SIVmac and MuLV vectors remains efficient (Hofmann et al. 1999). Recent data were shown, that the post-entry block is independent of the cell receptor used by the virus for entry, and was characterized by aberrant intracellular trafficking and impaired chromatin integration of HIV-1 (Cutino-Moguel and Fassati 2006). On the other hand, to-date all reports on lentiviral transgenic mammals—except for one with the EIAV virus (Whitelaw et al. 2004)—have used HIV-1 based vectors (for review Park (2007). Out of the five different ubiquitous promoters, which have widely been used in different lentiviral transgenic models, we chose the CAG promoter. The CAG promoter was used to create transgenic rabbits through traditional microinjection and an ubiquitously GFP expressing transgenic line was created (Takahashi et al. 2007). In our lentiviral construct, the SIV-based vector carries a central polypurine track (cPPT) and the post-transcriptional regulatory element of woodchuck hepatitis virus to increase transduction efficiency and transgene expression level (Negre and Cosset 2002).

The high pregnancy rate and large litter size, which we experienced, could be attributed to the reduced physical intervention of perivitelline space injection compared to pronuclear microinjection, which we did not take into consideration in the number of transferred embryos. In this exploratory study we transferred 20–24 embryos per recipient doe, which is more than a doe can normally take to term and nurse, therefore this resulted a high number of stillborns. It is important to note that the high number of stillborns was not due to or correlated with transgene integration/expression. It is possible therefore, that if fewer embryos were transferred per recipient female even greater overall efficiencies could have been achieved.

Although 6% GFP expressing transgenic founder rabbits is considerably higher than the 1% achieved by pronuclear microinjection, it is much lower than the reported figures—up to 97%—reached in transgenic porcine established both with HIV and EIAV lentiviruses (Hofmann et al. 2003; Whitelaw et al. 2004). This relatively low success rate could be the consequence of the low titre of our lentivirus compared to titres between 109 and 1010 TU/ml in the cases mentioned above. Nevertheless the low virus titre did not result, per se mosaic expression pattern. This is not the intrinsic property of the SIV-CAG-GFP vector either, since a transgenic rat founder created with the same batch of this lentivirus, revealed uniform GFP expression and the transgene inheritance followed Mendelian law through three generations (Bender et al. manuscript in preparation).

Western analysis underlined the quantitative variations in GFP protein expression among different founders and within the different organs of the same founder. The differential expression between the different organs was not the result of tissue specific epigenetic silencing, since we could show that the non-expressing tissues did not have integrated transgenic sequences. We assume that it is the consequence of delayed lentiviral integration in later developmental stages, with individual variance in individual rabbit embryos. The differential timing of pre-implantation development could partly explain the rabbit specific mosaic integration: it was published at first, half a century ago and confirmed recently, that rabbit embryos undergo rapid series of cell division and reach 4-cell stage within 11 h, contrary to rodent, sheep, swine and monkey embryos for which it takes 30–40 h following fertilization (Sultana et al. 2009; Witschi 1956). Recent data by Zaitseva et al. (2009) pointed out that while importin 7-depletion impaired HIV-1 function, other lentiviruses like HIV-2, SIV and EIAV were not affected. It was hypothetised that lentiviruses other than HIV-1 may have evolved to use alternative nuclear import receptors. Rapid early rabbit embryonic development combined with yet un-characterized rabbit-specific nuclear import receptors of the reverse transcription complex might result mosaic transgene expression, not reported in other species. We can not rule out the possibility that another non HIV-1 based lentiviral vector could have been more efficient in creating transgenic rabbit than SIV. Nevertheless our data clearly show that the SIV lentiviral vector based transgene is able to integrate and to be expressed in tissues derived from all three primary germ layers, in the male germ cells and in the extra-embryonic tissues of rabbit. Creation of transgenic rabbits and germ-line transmission of the transgene was achieved with an SIV-based lentiviral vector, in which the GFP marker gene was placed under the CAG promoter. In conclusion lentiviral transgenesis was successful in rabbit for the first time.

Acknowledgments

We thank Ariberto Fassati (University College, London, UK), Bruce Whitelaw and Bill Ritchie (Roslin, UK) and Marielle Afanassieff (INRA, Lyon, France) for the advices and helpful discussions. The authors thank G. Takács for the artwork. Supporting grants: OTKA T049034, GVOP-3.1.1.-2004-05-0071/3.0 and OM-00118/2008, OTKA PD78310.

Supplementary material

11248_2009_9356_MOESM1_ESM.tif (1.6 mb)
Supplementary Fig. 1. Integrated transgene copy number determination with Southern analysis. (A) Southern blot of BamHI-digested genomic DNA isolated from SIV-CAG-eGFP transgenic founder rabbit ear samples. Lines: 1–2: founder rabbit; line 3: non-transgenic rabbit, negative control; line 4: GFP-transgenic mouse (Kvell et al. 2009), positive control. (B) The lentiviral vector carrying the CAG-GFP transgene, LTR, long terminal repeat; CPPT, polypurine tract; WPRE, woodchuck hepatitis responsive element; dotted lines, rabbit genome. (TIFF 1652 kb)

Copyright information

© Springer Science+Business Media B.V. 2010

Authors and Affiliations

  • L. Hiripi
    • 1
  • D. Negre
    • 2
  • F.-L. Cosset
    • 2
  • K. Kvell
    • 3
  • T. Czömpöly
    • 3
  • M. Baranyi
    • 1
  • E. Gócza
    • 1
  • O. Hoffmann
    • 1
  • B. Bender
    • 1
  • Zs. Bősze
    • 1
  1. 1.Genetic Modification Program Group, Agricultural Biotechnology CenterGödöllőHungary
  2. 2.Inserm, U758, Human Virology Department Université de Lyon, Ecole Normale Supérieure de LyonLyonFrance
  3. 3.Department of Immunology and BiotechnologyUniversity of PécsPecsHungary

Personalised recommendations