ABSTRACT

Silencing of gene expression by RNA interference (RNAi) has become a widely used tool. For the study of mammalian gene function expression vectors for short hairpin RNA (shRNA) were developed. However the standard methods of shRNA transgenic (Tg) mice production have not been established. Sry (sexdetermining region on the Y chromosome) is a mammalian sex-determining gene on the Y chromosome. In mice, the transient expression of Sry in supporting cell precursor cells between 10.5 and 12.5 days postcoitus (dpc) triggers the differentiation of Sertoli cells from granulosa cells. Then high efficiency of Sry gene silencing in Tg mice should induce XY male-to-female sex reversal. An shRNA Tg mouse targeting Sry gene was attempted to be generated by pronuclear microinjection. A low rate (Tg pups/all pups born after microinjection = 2/154 to 7/178) of Tg pups was observed. These Tg mice showed no XY male-to-female sex reversal. The results suggest that exogenous expression of small RNA might exert a negative effect on embryonic development and another approach should be needed for RNAi transgenesis in mice.

1. INTRODUCTION

RNA interference (RNAi) has become a routine genetic tool to study gene function in mammalian cells including cancer cells [1]. Short hairpin RNA (shRNA) expression vectors have been developed which can be designed for knockdown of specific gene expression [2]. However the standard methods of shRNA transgenic (Tg) mice production have not been established. As RNA polymerase III (polIII) promoter such as U6, H1 and tRNA is ubiquitous promoter that is commonly used for shRNA expression, the phenotype of shRNA Tg mice should be similar to the phenotype of the knockout mice of the target gene. Sry (Sex determining region on the Y chromosome) is a transcription factor with a DNA-binding domain referred to as the high mobility group (HMG), which triggers a gene expression cascade required for initiating male sex differentiation in the bipotential indifferent gonads of mammals [3]. Mouse Sry is expressed for a brief period between 10.5 and 12.5 days post-coitus (dpc) in the supporting cells of undifferentiated gonads that differentiate into Sertoli cells instead of granulosa cells [4-6]. For specific purposes, shRNA targeting Sry was constructed [7] and the corresponding Tg mice were then attempted to be generated. As the pronuclear microinjection method is standard method for production of Tg mice, shRNA Tg mice were produced by pronuclear microinjection method in this study. If Sry knockdown in mice would be success, XY female-to-male sex reversal should be occurred. Then shRNA targeting Sry gene will be useful tool to establish the methods of RNAi transgenesis in mice.

2. MATERIALS AND METHODS

2.1. Animals

The following strains of mice were purchased from a commercial animal breeder (Sankyo Labo-Service Corporation, Inc., Tokyo, Japan): B6C3F1 (C57BL/6NxC3H/ He), C57BL/6J, and ICR. The mice were kept in an environment with regulated temperature (22˚C - 25˚C), humidity (40% - 50%), and illumination cycles (14-h light, 10-h dark), and were provided with food and water ad libitum. The present study was conducted according to the committee guidelines of the National Institute of Advanced Industrial Science and Technology for animal experiments and the Law for Prevention of Cruelty to Animals and the Guide for Care and Use of Laboratory Animals in Japan. The experiments complied with current Japanese laws.

2.2. Tg Mouse Generation

Tg mice were generated by microinjecting DNA into the pronuclei of zygotes collected from the oviducts of superovulated B6C3F1 females that were mated with B6C3F1 males. All methods for generating the Tg mice used here have been described in the protocol reported by Hogan et al. [8]. The construction of U6-shRNA1 (containing one shRNA expressing construct) and tRNAshRNA1-8 + Rz3 (eight shRNA and one ribozyme expressing constructs were connected) targeting Sry gene were described previously [7]. The EcoRI/HindIII DNA fragment for U6-shRNA1 and the EcoRI/MfeI DNA fragment for tRNA-shRNA1-8 + Rz3 were excised and separated by electrophoresis through 1% agarose gel; the fragment was then purified by CsCl ultra-centrifugation. The purified DNA fragment was dissolved in a solution containing 10 mM Tris-HCl (pH 7.4) and 0.25 mM EDTA (pH 7.4) and was used for pronuclear microinjection. To identify Tg animals and sex chromosome karyotypes (XX or XY), genomic DNA was isolated from the tip of the tail, and the genomic DNA was screened by polymerase chain reaction (PCR) amplification as previously described [9]. Transcripts of the Sry in the XY gonads at 11.5 dpc were determined by reverse transcription polymerase chain reaction (RT-PCR) as previously described [9].

2.3. Histology

For the histological analysis, tissues were fixed in 4% paraformaldehyde. After fixation, the tissues were processed for paraffin embedding as previously described [6]. The tissues were then sectioned at 6 μm, and the sections were used for hematoxylin and eosin staining.

3. RESULTS

Production of Tg Mice

The constructs of shRNA targeting Sry gene (U6-shRNA1 or tRNA-shRNA1-8 + Rz3) were used for DNA microinjection to produce Tg mice (Figures 1(a) and (b)). The constructs were selected since these constructs were exhibited the highest silencing effect toward Sry [7]. In the analyses of DNA pups born, the percentages of Tg mice were 3.9% (Tg pups/all pups born after microinjection = 7/178) for U6-shRNA1 and 1.3% (Tg pups/all pups born after microinjection = 2/154) for tRNA-shRNA1-8 + Rz3. Seven U6-shRNA1 Tg lines (5 female and 2 male) were analyzed. All Tg mice were fertile. The female karyotype was XX, and that of the male was XY. F₁ XY Tg mice from breeding with the all five F₀ XX female Tg mice were male (Figure 1(c)). Two tRNA-shRNA1-8 + Rz3 Tg lines (female and male) were analyzed. The female karyotype was XX, and that of the male was XY. All Tg mice were fertile. F₁ XY Tg mice from breeding with the F₀ XX female Tg mouse were male. There was no histological abnormality in the testis of the Tg mice (Figures 1(e) and (f)). Sry expression levels of Tg XY gonads at 11.5 dpc were not lower than that of wild-type littermates (Figure 1(d)). Thus, shRNA was silenced at the inserted locus and not expressed in Tg tissues or Sry expression was not suppressed by the shRNA. Then, male-to-female sex reversal could not occur with the polIII promoter driven constitutive shRNA expressing transgene by pronuclear microinjection method.

4. DISCUSSION

In this study, an shRNA Tg mouse production approach was used to repress the expression of Sry. Foreign mouse Sry has been shown to induce XX female-to-male sex reversal [3]. Then it was expected that transgenesis of shRNA targeting Sry would induce XY male-to-female sex reversal. Tg mice expressing shRNA targeting Sry

were attempted to be generated. The efficiency of transgenesis was remarkably low (1.3% to 3.9%) compared with the efficiency of usual Tg experiments [10-13]. Furthermore, no XY male-to-female sex reversal occurred with the transgene. All the female Tg mice were characterized as XX and the male Tg mice were characterized as XY. This result might suggest that ubiquitous and strong expression of shRNA yields embryonic lethality; the integration of the transgene at the locus where the transgene is silenced could explain why the XY sex reversal was not occurred in the Tg mice. It is also possible that the design of shRNA sequences was not suitable for suppression of Sry expression. However the shRNA constructs exhibited the strong silencing effect toward Sry gene in the cultured cells [7]. The knockdown efficiency of the shRNA constructs in vivo may be confirmed by organ culture on agar blocks and magnetically induced transfection using embryonic genital ridges [14] or by injection into tail vein of pregnant mice [15].

For transgenic RNAi in mice, one-cell embryos [16-29] or embryonic stem (ES) cells [30-37] have been used as the target (Table 1). The methods of ES cells need several months for many steps such as vector construction, clone isolation, chimera mouse production and breeding. However ES cells have made possible a ubiquitously active locus (e.g., ColA1 and Rosa26) targeting transgene integration [34-37]. One-cell embryos have been used for pronuclear microinjection [16-25] or lentivirus infection [17,26-29]. Long double strand RNA (dsRNA) (around 500bp) expressing vectors were constructed for microinjection [19-23]. However these constructs were effective only oocytes or early embryos as expression of dsRNA induces apoptosis by dsRNA-dependent protein kinase (PKR) in most mammalian cell types [38]. In the case of shRNA expressing constructs, Carmell et al. reported that Tg mice produced by pronuclear microinjection did not show a phenotype that was expected for the targeted genes such as tyrosinase (albino), myosin VIIa (shaker), Bmp-5 (crinkled ears), Hox a10 (limb defects), homogentisate 1,2-dioxygenase (urine turns black upon exposure to air), Hairless (hair loss) and melanocortin 1 receptor (yellow) [17]. The report agrees with the present study in that pronuclear microinjection is not suitable for RNAi transgenesis. On the other hand Peng et al. reported that pronuclear microinjection of shRNA-expressing constructs was used for genetic screening in mice. In the study one of the transgene targeting p57 was effective and the knockdown effect was transmitted for many generations [18]. Then it might depend on the targeting gene whether the shRNA-expressing constructs have knockdown effect in the Tg mice produced by pronuclear microinjection. The method of lentivirus infection has some advantages over the pronuclear microinjection method [26]. First, the injection of the lentivirus into the perivitelline space is less invasive than the pronuclear microinjection. Second, the efficiency of transgenesis is much higher than that of pronuclear microinjection. Third, single copy insertions are occurred by lentivirus infection. On the other hand multiple copies are inserted into one locus in the most cases of transgenesis by the pronuclear microinjection method. Single copy insertion is attractive for RNAi transgenesis since it is expected that too high expression of shRNA should induce embryonic lethality in this study. Moreover high efficiency of transgenesis would make it easy to analyze F₀ Tg embryos and it should be useful tool when the

gene knockdown causes embryonic lethality. F₀ Tg embryos also can be analyzed using tetraploid aggregation technique and ES cells containing shRNA-expressing construct [30-33]. Kirilov et al. reported that lentivirus mediated F₀ Tg mice were genetically mosaic and the mosaicism led to a reduced rate of germ line transmission [29]. It is possible that the overexpression of shRNA might have interfered with the processing of endogenous micro RNA (miRNA) and should be cause of the high mosaicism and the low degree of germ line transmission.

To overcome the problem, inducible shRNA expression system should be useful such as Cre/loxP system [24,32, 33] and tetracycline system [25,34-37]. In the Cre/loxP system, a loxP-flanked stuffer sequence should be inserted into the shRNA expression construct and can be removed by Cre. After the recombination, shRNA can be expressed and work for target gene knockdown [24,32, 33]. Tissue and stage specific promoter can be used for the control of Cre expression and the Cre Tg mouse will be used for the spatiotemporally controlled shRNA expression. Chemical inducible Cre recombination system also has been developed using doxycycline (a member of the tetracycline antibiotics) dependent transcription or tamoxifen (an agonist/antagonist of the estrogen receptor) dependent nuclear translocation. The tetracycline system should need two Tg lines—the tetracycline transactivator (tTA; for Tet-Off) or reverse tTA (rtTA; for Tet-On) expressing line and the line containing the construct of shRNA under the control of tetracycline responsive promoter. Dickins et al. established Tg lines expressing miRNA based shRNA controlled by tetracycline responsive–cytomegalovirus promoter using pronuclear microinjection method, and the doxycycline-dependent reversible induction was worked effectively [25]. More-

over, tissue specific polII promoter would be applied for the tetracycline-dependent shRNA expression system [25]. However, Cre/loxP system might be more useful than tetracycline system since strong transgene expression can be switched on by Cre/loxP system when weak polII promoter such as Sry promoter would be used [6, 39,40]. As blood-brain barrier and blood-testis barrier have low permeability to doxycycline, the tetracycline inducible system may not be applicable for the expression in brain and testis [41]. As for the knockdown of Sry gene at the embryonic stage the tetracycline inducible system should be applicable since shRNA expression was induced in Tg embryos when the water or food pellets containing doxycycline were supplied to the pregnant mothers [35,36]. In conclusion the pronuclear microinjection method is not suitable for the generation of constitutive RNAi Tg mice for Sry gene knockdown. Recently, methods of conditional RNAi transgenesis in mice have been developed [42,43]. Figure 2 shows flowchart for selection of an appropriate RNAi transgenesis method. As male-to-female sex reversal is one of the most obvious phenotype, Sry is useful as a target gene to test the RNAi transgenesis method.

5. ACKNOWLEDGEMENTS

This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan.

REFERENCES