One-step Cre-loxP organism creation by TAx9

Abstract

The creation of organisms with Cre-loxP conditional gene recombination systems often faces challenges, particularly when creating the initial (F0) generation with both a Cre recombinase and a DNA site flanked by loxP elements (floxed site). The primary reason is that it is difficult to synthesize a single plasmid with both the Cre gene and the floxed site, since Cre-mediated recombination spontaneously occurs when the plasmid is amplified in Escherichia coli bacterial cells. Here, we introduce an artificial nucleic acid sequence TATATATATATATATATA, named TAx9, that enables the integration of both the Cre gene and the floxed site into a single plasmid. TAx9 effectively blocks spontaneous Cre-mediated recombination in E. coli cells. Using this system, we created an F0 generation of transgenic newts and CRISPR-Cas9 knock-in mice with tissue-specific Cre recombination triggered by tamoxifen. TAx9 technology will be a powerful strategy for creating organisms capable of conditional genetic modification in the F0 generation, accelerating various life science research by reducing the time and cost for ultimately establishing and maintaining lines of genetically modified organisms.

Fig. 1. TAx9 guards a single Cre-loxP plasmid from Cre-mediated recombination in E. coli cells.

a The plasmid used to optimize the TA repeat element. Each of the DNA elements listed with their ID was inserted into the TA(ID) site to examine its shielding effect on Cre-mediated recombination in E. coli cells. Here, WT is the E. coli LexA binding motif, i.e., the SOS box. To detect deletion of the floxed site due to Cre-mediated recombination, PCR primers were set on the CAGGs promoter and the site adjacent to the mCherry 5′ region to amplify a 2.5 kbp (intact) or 1.5 kbp (recombined) region. 2× HS4: a double core of the HS4 insulator. Each of the EGFP, mCherry, and CreER^T2 gene cassettes was attached with an eukaryotic terminal polyadenylation signal sequence on its 3′ end. pA: rabbit β-globin polyadenylation signal; SpA: SV40 polyadenylation signal. b Evaluation of recombination in the primary E. coli cells following transformation. Colony PCR was carried out using single colonies that were randomly selected from an LB plate cultured at 30 °C for 18 h. Lane numbers indicate colony ID. Cre-mediated recombination occurred in all colonies except for those of ATx8 and TAx9 (asterisks). Pink arrowheads: 1.5 kbp (recombined). M: size marker. c Integrity of the plasmids after the culture of E. coli cells in LB medium. Here, the E. coli cells containing the plasmids with ATx8 (clone ID: 1–3 in b) and those with TAx9 (clone ID: 1–6 in b) were further cultured in three and six tubes, respectively, for 18 h at 30 °C. The plasmids were purified from the cultures and digested with the I-SceI. The sizes of the I-SceI-flanked region and the plasmid backbone were 9.5 kbp and 2.9 kbp, respectively. Lane numbers indicate tube ID. Cre-mediated recombination (8.5 kbp; arrowheads) was detected with ATx8 (asterisks) but was never detected with TAx9. Note that the band of recombined plasmids in lane #3 of ATx8 was faint compared to that in lane #1. M1 and M2: size markers.

Fig. 2. The E. coli LexA repressor protein binds to the TAx9 element.

A gel shift assay was performed to examine the binding affinity of LexA proteins to the TAx9 element in comparison to the LexA binding motif (WT). a In the presence of WT oligonucleotides (duplex DNA) at a concentration of 10 pmol, synthetic LexA proteins formed complexes with them at 5 pmol and above. On the other hand, in the presence of 10 pmol of TAx9 oligonucleotides (lower panel), synthetic LexA proteins formed complexes with them at 20 pmol. b Increasing the concentration of TAx9 oligonucleotides to 20 pmol resulted in a more distinct band of complexes at 20 pmol of the synthetic LexA proteins. Note that in both conditions (a, b), free TAx9 oligonucleotides decreased as the concentration of synthetic LexA protein increased; the data using GCx10 oligonucleotides is shown as a negative control (lower panel in b). Taken together, these results indicate that under the current experimental conditions, LexA proteins bind to TAx9 oligonucleotides, although the binding affinity of LexA proteins to TAx9 oligonucleotides is much lower than that to WT oligonucleotides. Black arrowheads indicate the location of complexes.

Fig. 3. TAx9 enables the creation of inducible Cre-loxP RPE cell-labeled newts at F0.

a Schematic diagram showing the plasmid pmCherry[EGFP]<CAGGs-TAx9-cpRPE65>CreER^T2(I-SceI). b Outline of the transgenic protocols. 4-OHT: 4-OH tamoxifen. c Screening by fluorescence at an early blastula stage (St. 10) and at the larval stage (St. 43). Individuals never exhibited mCherry fluorescence, and instead emitted EGFP fluorescence. The differential intensity of EGFP fluorescence is thought to be caused by the number of transgene cassettes inserted in the genome, or their location. In St. 10, white, pink, and yellow arrows indicate strong, average, and weak EGFP fluorescence, respectively. Embryos exhibiting strong (white broken circle) or average (pink broken circle) EGFP fluorescence were screened. d Screening by genomic PCR at the swimming larval stage (St. 59). Lane numbers indicate the ID of the individual. P: plasmid DNA (control). M: size marker. The full site (6.5 kbp), driver site (1.9 kbp), and floxed site (2.5 kbp) were examined (see a). A 0.88 kbp site of the tyrosinase gene was also examined as the positive control. Black arrowhead: nonspecific band. Pink arrowhead: 1.5 kbp or 5.5 kbp, the size of the band that would be detected if Cre-mediated recombination had occurred. In this group, all larvae were screened as positive, except for the ID#1 individual, whose full and floxed sites were not amplified by PCR. Undesired Cre-mediated recombination was never recognized. e Fluorescence after 4-OHT administration. Individuals at St. 59, just before metamorphosis, were treated twice with 4 µM 4-OHT (see b). mCherry fluorescence was never detected, at least not on the surface of the body, at 48 h post-treatment (upper panels) or even beyond metamorphosis (lower panels). All of the metamorphosed individuals (n = 10) grew normally, beyond 1 year. Scale bars: c 2 mm for St. 10, 2 mm for St. 43; e 0.5 cm for St. 59, 1 cm for 1 year.

Fig. 4. 4-OHT induces the expression of mCherry in RPE cells.

a Representative images showing 4-OHT-dependent induction of mCherry expression in RPE cells. Larvae at St. 59 were treated with either 4-OHT (+4-OHT) or the solvent (DMSO alone) (−4-OHT) at 48 h before obtaining eyeballs (three larvae for each group). Here, to visualize immunoreactivity in the RPE layer, the ABC-DAB method was applied and melanin pigments were bleached. Arrowheads point to mCherry-immunoreactive cells along the RPE layer. b Representative images showing the double labeling of RPE65 and mCherry in RPE cells by antibodies (three larvae). DAPI: nuclear stain. The lower panels are enlarged views of the area enclosed by the dashed rectangles in the upper panels. In the lower panels, dashed lines highlight the RPE layer. White arrowheads point to the RPE cells with both RPE65- and mCherry-immunoreactivity. GCL: ganglion cell layer; INL: inner nuclear layer; ONL: outer nuclear layer. Scale bars: a, 100 μm; b, 200 µm for upper panels, 100 µm for lower panels.

Fig. 5. TAx9 enables the creation of inducible Cre-loxP SMF cell-labeled mice at F0.

a Schematic diagram showing the plasmid pmCherry[EGFP]<CAGGs-TAx9-hACTA1>CreER^T2(ROSA26 Arms) designed to create SMF cell-labeled mice by the CRISPR-Cas9 knock-in (KI) protocol targeting the ROSA26 locus^, (also see Supplementary Fig. 8). The sites amplified by genomic PCR are noted. b–d Screening of KI individuals by genomic PCR. Here, a total of 76 one-month-old individuals who survived beyond the weaning period were examined. Lane numbers indicate the ID of the individual. P: plasmid DNA (control). M: size marker. The first PCR for the 3′ arm crossing site (3.3 kbp) (b) followed by the second PCR for the 5′ arm crossing site (5.2 kbp) (c) screened nine individuals as positive candidates (white asterisks). Individuals with ID#s 13, 22, 30, 37, 43, 44, 59, and 60 are considered to be individuals with DNA fragments containing the 3′ arm crossing site but not the 5′ arm crossing site, that were inserted randomly into the genome. KI of the single Cre-loxP construct in all these candidates was confirmed by the last PCR for the driver site (2.9 kbp), the floxed site (2.5 kbp), and the full site (7.5 kbp) (d). e Fluorescence of the tail tip of KI-positive (#7, #8, #9) and KI-negative (#6) individuals. All of the KI-positive individuals showed EGFP fluorescence but not mCherry fluorescence, indicating that Cre-mediated recombination had not occurred. Pink arrowhead: 4.2 kbp (c), 1.5 kbp (d), or 6.5 kbp (d), indicating the size of the band that would be detected if Cre-mediated recombination had occurred. Scale bar: 5 mm.

Fig. 6. Tamoxifen induced the expression of mCherry in SMF cells.

a Genomic PCR for the floxed site in F1 heterozygous KI mice (intact, 2.5 kbp; recombined, 1.5 kbp). These mice were injected with three doses (3D) or six doses (6D) of tamoxifen or the solvent corn oil alone (TAM−) two weeks before limb sample collection (n = 3 each). Lane numbers indicate the ID of the individual. WT: genomic DNA of a wild-type C57BL/6J mouse. P: plasmid DNA (control). M: size marker. Pink arrowheads: 1.5 kbp (recombined). b−e Representative images showing tamoxifen-dependent induction of mCherry expression in SMF cells (n = 3 for each). F1 heterozygous KI mice were injected with either six doses of tamoxifen (TAM+) or the solvent corn oil alone (TAM−). The limb samples were cross-sectioned in the middle of the zeugopod (20 µm thick). mCherry fluorescence was detected in the TAM+ mice (b, c) but not in the TAM− mice (d, e). Note that EGFP fluorescence in the TAM+ mice was weaker than in the TAM− mice, suggesting recombination of the floxed site. The optical sections (1.3–2.0 µm thick) of the boxed areas of the merged images in (b) and (d), acquired by laser confocal microscopy, are shown in (c) and (e), respectively. Scale bars: b and d 1 mm for upper panels, 200 µm for lower panel; c and e 100 µm.