A simple way to generate cell-specific knockdown mice by double transfection of Cre/loxP lentiviral vector

Abstract

Background: In case of eternal gene manipulation, there can be limits in figuring out functions of genes that are temporarily activated at diseases. To overcome such problems, cell-specific and tissue-specific gene technology was recently developed and has been greatly helpful for understanding functions of genes of persons suffering from various diseases. It is one of organism-specific gene expression manipulation systems widely utilized for conditional gene knockout in selective cell or tissue by Cre, a site-specific recombinase. To develop gene manipulation animals, Cre genetically modified animals are established through fertilization and crossbreeding after cell-specific Cre genes are injected to embryonic stem cells. However, there are limits to use it in terms of time and money. On the other hand, lentivirus-based gene delivery system, gene has incorporated into host chromosomes and continuously expressed for a long time, therefore, genes can be effectively conveyed to already differentiated and non-proliferating cells such as nerve cells.Purpose: In this study, I invented a simple way to knockdown a specific gene in a cell-specific pattern in adult mice by lentivirus-assisted transfer of shRNA.Methods: In vitro, for lentiviral transfection, 4 × 105 TU of the lentivirus suspension containing LV-Hoxb7 Cre and/or LV-shAQP3 was added to cultured Mouse renal collecting duct cells and mouse mesangial cells. After 48 hr, the media were changed to routine culture media. At 72 hr after the media change, cells were harvested. In vivo experiments, there were three models to check efficiency of each virus. First, LV-Hoxb7 Cre is injected into the loxP-EGFP mice to check efficiency of the Hoxb7 promoter. These mice were divided into two groups, and were injected with PBS (Con) or LV-Hoxb7 Cre. For the second animal experiments, Hoxb7 Cre transgenic mice were used. They were classified into three groups based on what was injected with PBS (Con), LV-loxP scr, or LV-loxP shAQP3. For the third animal experiments, male C57BL6/J mice were assigned to one of the four groups, and were injected with PBS (Con), LV-Hoxb7 Cre, LV-loxP shAQP3, or LV-Hoxb7 Cre + LV-loxP shAQP3. All animals were given the selected treatment on days 0, 4, and 7 via hydrodynamic tail vein injection as described previously. For C57BL6/J mice treated with consecutive injection of LV-Hoxb7 Cre and LV-loxP shAQP3, LV-shAQP3 was introduced on days 3, 7, and 10 after the final injection of LV-Hoxb7 Cre. Briefly, a volume of 1 ml of PBS alone or PBS containing 4 × 108 TU of lentivirus were injected to mice via the tail vein within 10 sec. Mice were sacrificed at 6-week after the first injection of PBS or lentivirus. EGFP, mCherry, and AQP3 protein expression in renal collecting duct cells, the kidney were evaluated by Western blot analysis, immunofluoresence staining, and AQP3 mRNA expression by real-time PCR.Results: In vitro, LV-Hoxb7 Cre, which contained Hoxb7 promoter, worked only in collecting duct cells due to the presence of Hoxb7 in collecting duct cells but not in mesangial cells. Furthermore, combined infection of collecting duct cells with LV-Hoxb7 Cre and LV-loxP shAQP3 significantly inhibited the protein expression of AQP3 along with the disappearance of EGFP protein expression, suggesting that LV-Hoxb7 Cre and LV-loxP shAQP3 used in the present study worked together effectively.In vivo, both EGFP and mCherry protein were completely expressed at the same site, in kidney collecting duct cells of loxP-EGFP mice, where Hoxb7 promoter was working, but not in the liver of LoxP-EGFP + LV-Hoxb7 Cre mice. The protein expression of EGFP in the kidney was observed in Hoxb7 Cre mice injected with LV-loxP scr or LV-loxP shAQP3. Furthermore, AQP3 protein and mRNA expression were significantly reduced in Hoxb7 Cre mice injected with LV-loxP shAQP3 compared to inject with PBS and LV-loxP scr. In addition, even though urine volume was significantly increased, urine osmolality was significantly decreased by 44.4% in Hoxb7 Cre + LV-loxP shAQP3 mice compared to the other groups. In contrast, there was no significant change in the protein expression of AQP3 in the colon and trachea by LV-loxP scr or LV-loxP shAQP3 injection in spite of demonstrable expression of EGFP protein. Finally, I tried to generate AQP3 knockout mice in C57BL/6J mice by consecutive injection of LV-Hoxb7 Cre and LV-loxP shAQP3. When LV-Hoxb7 Cre was introduced in these mice, the expression of mCherry protein was found only in the kidney collecting duct cells. Moreover, injection of LV-loxP shAQP3 resulted in an increase in EGFP protein expression in all kidney cells except for collecting duct cells. However, there were no significant differences in the protein and mRNA expression of AQP3 in C57BL/6J mice injected with LV-Hoxb7 Cre or LV-loxP shAQP3 alone. In contrast, when consecutive injection of LV-Hoxb7 Cre and LV-loxP shAQP3 was performed in these mice, AQP3 protein and mRNA expression were significantly reduced. Even though urine volume was significantly increased, urine osmolality was significantly decreased in LV-Hoxb7 Cre and LV-loxP shAQP3 injected mice compared to the other groups. Meanwhile, the protein expression of EGFP was also observed in the colon and trachea of C57BL/6J mice by the administration of LV-loxP shAQP3 with or without LV-Hoxb7 Cre, but AQP3 protein expression was not changed by these treatment.Conclusions: These findings suggest that double transduction of Cre- and loxP-based lentivirus can be a simple way to generate cell-specific knockdown mice, and this method may also be applicable to other species.