Animals
We have complied with all relevant ethical regulations for animal use.
Adult C. pyrrhogaster newts (total body length: males, ∼9 cm; females, 11–12 cm) of the Toride-Imori line23 were used to obtain fertilized eggs for transgenesis. They were originally captured by a provider (Mr. Kazuo Ohuchi, Misato, Saitama, 341-0037 Japan; http://kaeru-kerokero.la.coocan.jp/index.html) within a ∼25 km diameter around the Miyayama area (35.130013, 140.013842) in Kamogawa city, Chiba prefecture, Japan. No specific permissions were required for the location of capture. Newts were reared for longer than one year in the laboratory (University of Tsukuba) by keeping them in plastic containers in which a shallow layer of water (∼3 cm deep) and a resting island were placed, at 18°C and in natural light, as was previously described24. In this study, 30 females and 10 males were kept in separate containers for 60 days until the transgenic experiments began. Both of them were fed frozen mosquito larvae (Akamushi; Kyorin Co., Ltd., Hyogo, Japan) three times a week. Containers were kept clean. To obtain fertilized eggs at the 1-cell stage (F0), adult newts were maintained according to a previously established protocol10. Using fertilized eggs (n = 84), transgenesis was carried out (see below), and embryos and larvae were staged according to established criteria24. A total of 10 transgenic individuals were finally screened and their recombination was examined without distinguishing between sexes (see the “Induction of Cre-mediated recombination” section).
Adult C57BL/6J mice (Strain #: 000664; RRID: IMSR_JAX:000664) were purchased from Jackson Laboratory Japan Inc. (Kanagawa, Japan)11 and kept in the Laboratory Animal Resource Center in the Transborder Medical Research Center. They were housed with a 14-h light: 10-h dark cycle. Air quality was maintained under specific pathogen-free conditions (23.5 °C ± 2.5 °C and 52.5% ± 12.5% relative humidity)11. They were fed a commercial diet (MF diet; Oriental Yeast Co., Ltd., Tokyo, Japan) and could drink filtered water, both of which were freely accessible. For CRISPR-Cas9 KI, a total of 283 embryos were used, a total of 76 individuals (males: 39; females: 37) survived beyond the post-weaning period, and a total of nine individuals (males: two; females: seven) were finally screened as F0. For experiments with F1 to F4, a total of 117 individuals (males: 57; females: 60) were used. In the recombination experiments in Fig. 6, three males and three females were used for the test (for 3D, one male, two females; for 6D, two males, one female) while one male and two females were used for the control (TAM−) (see the “Induction of Cre-mediated recombination” section).
Only skilled researchers handled both the newts and mice. There were no unexpected adverse events.
Plasmid construction
All plasmids used herein were constructed and synthesized by standard molecular cloning techniques. In brief, to screen for optimal DNA elements to protect Cre-loxP single plasmids from Cre-mediated recombination in E. coli cells, a region containing CarA>CreERT2 was removed from the plasmid pmCherry[EGFP]<CAGGs-CarA>CreERT2 (I-SceI)25 with the restriction enzymes, BspEI and Ajul, and then a fragment BspEI-PpuMI-I-PpoI>CreERT2 was inserted back in, creating a new intermediate plasmid pmCherry[EGFP]<CAGGs-BspEI-PpuMI-I-PpoI>CreERT2 (I-SceI). The I-PpoI restriction enzyme recognition site was used to insert a promoter to drive the expression of CreERT2. In this study, a 1.1 kbp fragment containing the cpRPE65 promoter and its 5′ UTR region9 were inserted. The BspEI-PpuMI site was referred to as TA(ID), which was used to insert a candidate protective DNA element indicated in Fig. 1a.
To create inducible Cre-loxP RPE cell-labeled newts, the plasmid pmCherry[EGFP]<CAGGs-TAx9-cpRPE65>CreERT2 (I-SceI) was constructed (Fig. 3a). Here, TAx9 was inserted at the TA(ID) site. The transgene cassette was flanked by the I-SceI meganuclease recognition sites. The chicken β-globin HS4 2× core insulators (a gift from Dr. Gary Felsenfeld at the National Institute of Health, Bethesda, MD, USA) were placed at both ends of the Cre-driver cassette and the floxed reporter cassette to reduce the positional genomic effects and cis interactions within the transgene. Each of the EGFP, mCherry, and CreERT2 gene cassettes was followed by an eukaryotic terminal polyadenylation signal sequence (for details, see Fig. 1).
To create inducible Cre-loxP SMF cell-labeled mice, the plasmid pmCherry[EGFP]<CAGGs-TAx9-hACTA1>CreERT2(ROSA26 Arms) was constructed (Fig. 5a and see Supplementary Fig. 8). At first, the human ACTA1 skeletal muscle promoter4 fragment (−1919 to +156 in exon 1) was inserted at the I-PpoI site in the plasmid pmCherry[EGFP]<CAGGs-TAx9-I-PpoI>CreERT2 (I-SceI), creating the plasmid pmCherry[EGFP]<CAGGs-TAx9-hACTA1>CreERT2 (I-SceI). Next, the entire transgene cassette was released by I-SceI digestion and inserted into a ROSA26 targeting vector11,12, so that the transgene cassette had the homologous arms targeting the ROSA26 locus adjacent to both ends of the cassette (1.1 kbp at the 5′ side and 2.8 kbp at the 3′ side)11,12. This cassette also contained three sets of the chicken β-globin HS4 2× core insulators.
For all plasmids, E. coli cells of the NEB stable strain (C3040I; New England Biolabs, Ipswich, MA, USA) were transformed and cultured on LB plates at 30°C for 18 h. Colonies on the plates were individually examined by PCR for Cre-mediated recombination in plasmids, and cultured in LB medium for an additional 18 h at 30°C. The plasmids synthesized in the E. coli cell culture were purified using the GeneJET Plasmid Miniprep Kit (K0502; Thermo Fisher Scientific Inc., Tokyo, Japan) or the PureLink HiPure Plasmid Filter Midiprep Kit (K210014, Invitrogen; Thermo Fisher Scientific Inc.) according to the manufacturer’s instructions.
Exceptionally, as explained in the “Attaining the TA repeat sequences” section of the Supplementary Information, insert DNA was obtained by PCR and ligated into plasmids using the In-Fusion method (Z9649N; In-Fusion® HD Cloning Kit, Takara Bio Inc., Shiga, Japan) according to the manufacturer’s instructions. E. coli cells used for transformation and culture were the Stbl3 strain (One Shot™ Stbl3™ Chemically Competent E. coli; C737303, Invitrogen, Thermo Fisher Scientific Inc.). The culture conditions were the same as those for the NEB stable strain mentioned above. As shown in Supplementary Fig. 7, inserts containing TA/AT-rich sites were amplified using Tks Gflex DNA Polymerase (R060A; Takara Bio Inc.) according to the manufacturer’s instructions. In the experiments to evaluate Cre gene expression in E. coli cells (Supplementary Fig. 5), total RNA was purified using the SV Total RNA Isolation System (Z3100; Promega, Madison, WI, USA) and amplified using the One Step RNA PCR Kit (AMV) (RR024A; Takara Bio Inc.) according to the manufacturer’s instructions.
Gene modification
For transgenic newts, an I-SceI protocol we established for C. pyrrhogaster10 was applied. Briefly, fertilized eggs (one-cell stage embryos) were individually micro-injected with 100 pg of plasmid DNA and 2.0 × 10-3 U of I-SceI, which were dissolved in 1× I-SceI buffer. For CRISPR-Cas9 KI mice, a protocol we established11,12,13 was applied with a guide RNA targeting 5′-CCAGTCTTTCTAGAAGATGGGCG-3′ in the ROSA26 locus (chr6: qE3). Preparation of crRNA, tracrRNA, Cas9, and donor DNA, micro-injection into mouse zygotes, and embryo transfer into pseudopregnant mice were all performed using the same methods as in our previous papers11,12,13.
Anesthesia
Larval newts (St. 59) were anesthetized with FA100 (4-allyl-2-methoxyphenol; DS Pharma Animal Health, Osaka, Japan) dissolved to 0.05% (v/v) in filtered tap water at 18 °C; for image capture, 30 min; for tail tip sampling, 30 min; for head collection or euthanasia for culling, deep anesthesia as long as 60 min (the larvae remain in an upside down position and do not respond to cutaneous pinching with forceps) followed by decapitation. Mice were anesthetized by intraperitoneal injection of a mixture of ketamine and xylazine; for tail tip sampling, 80 mg/kg ketamine and 10 mg/kg xylazine (10 min); for forelimb collection or euthanasia for culling, an overdose of the anesthetics (triple the dosage; 15–20 min) followed by cervical dislocation. In both conditions, we confirmed that the mice were unresponsive to pinching of their legs and tails.
Genomic PCR
Genomic DNA was obtained from the tail tips of larval newts (St. 59) and mice (1 month old) with a Wizard Genomic DNA Purification Kit (A1120; Promega). For mice, limb muscles were also used (see below). The genomic DNA solution was aliquoted to 10 ng/µL and kept at 4 °C. PCR was carried out using Tks Gflex DNA Polymerase (R060A; Takara Bio Inc.) according to the manufacturer’s instructions. Primer sets are listed in Supplementary Table 1.
Induction of Cre-mediated recombination
For larval newts (St. 59), the method we established26 was applied. In brief, larvae were incubated in filtered tap water containing 4 μM 4-OHT ((Z)-4-hydroxytamoxifen, H7904-5MG; Sigma-Aldrich; Merck, Darmstadt, Germany) and 0.8% (v/v) dimethylsulfoxide (DMSO; 045-07216; FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) for 24 h at 22 °C in the dark (first round). They were immediately transferred into fresh 4-OHT-containing water and incubated for an additional 24 h in the same conditions (second round). For control, larvae were incubated in filtered tap water containing DMSO alone.
For mice (1 month old), a standard peritoneal injection protocol was applied. In brief, tamoxifen (T5648-1G, Sigma-Aldrich; Merck) was dissolved in corn oil (032-17016, Wako Pure Chemical Corporation) at a concentration of 20 mg/mL for 1 h at 65 °C in the dark and vortexed, and the liquid was divided into 1 mL aliquots and stored at 4 °C (it was used within 1 week). Just before injections, aliquots were prewarmed for 10 min at 65 °C, cooled to room temperature, and then injected with a 27-gauge needle (NN-2719S; Terumo, Tokyo, Japan) to achieve 150 mg/kg body weight. Injections were given once a day for three consecutive days, followed by a four-day interval, and then again for three consecutive days (six doses in total). Following the last injection, mice were allowed to rest for 2 weeks. For the control, corn oil alone was injected.
Tissue preparation
In this study, a fixative, 3% glyoxal27,28 (078-00905, FUJIFILM Wako Pure Chemical Corporation) and 2% paraformaldehyde (PFA; Catalog 26126-25; Nacalai Tesque, Inc., Kyoto, Japan) in phosphate-buffered saline (PBS; pH 7.0, adjusted with acetic acid) was prepared, stored in an amber-colored glass bottle at 4°C, and used within 1 month.
Heads of larval newts (St. 59) were collected in PBS on ice, and immediately transferred into the fixative and incubated at 4 °C for 8 h. The samples were washed several times for 3 h each in PBS at 4 °C, decalcified in 10% ethylenediamine-N,N,N′,N′-tetraacetic acid (EDTA) in PBS (pH 7.0, adjusted with NaOH) for 48 h at 4°C, rinsed in PBS, and then transferred to 30% sucrose in PBS at 4 °C.
Forelimbs of mice (1 month old) were collected in PBS on ice following cervical dislocation, their skin was removed, and they were trimmed to obtain a portion from the wrist to the distal end of the humerus. Limb muscles for genomic DNA were sampled at this time. The deskinned forelimb samples were immediately fixed at 4 °C for 18 h, washed several times for 3 h each in PBS at 4 °C, and incubated in a decalcifying solution (10% EDTA in PBS (pH 7.0, adjusted with NaOH)) for 72 h at 4 °C, rinsed in PBS, and then transferred to 30% sucrose in PBS at 4 °C.
Following the equilibration process in sucrose solution, newt head and mouse forelimb samples were embedded into an O.C.T compound (45833; Sakura Finetech, Tokyo, Japan) and sectioned to 20 µm thickness using a cryotome (CM1860 UV, Leica Biosystems, Tokyo, Japan). Frozen sections were air dried in the dark for 24 h before proceeding to the direct observation of their fluorescence or immunohistochemistry.
Immunohistochemistry
Primary antibodies were mouse monoclonal anti-RPE65 antibody (1:500; MAB5428; EMD Millipore, Burlington, CA, USA), and rabbit polyclonal anti-RFP antibody (for mCherry; 1:500; 600-401-379; Rockland Immunochemicals, Pottstown, PA, USA). Secondary antibodies were Alexa 488-conjugated goat anti-mouse IgG (H+L) antibody (1:500; A11001; Thermo Fisher Scientific Inc.); rhodamine (TRITC)-conjugated affiniPure goat anti-rabbit IgG (H+L) antibody (1:500; Code: 111-025-003; Jackson ImmunoResearch Laboratories, West Grove, PA, USA), and biotinylated goat anti-rabbit IgG (1:250; BA-1000; Vector Laboratories, Inc., Newark, CA, USA). Immunohistochemistry was performed as in a previous paper26, with some modifications. In brief, for immunofluorescence labeling, tissue sections were washed (PBS, 15 min; 0.1% Triton X-100 in PBS, 15 min; and PBS, 15 min), incubated with blocking solution (2% normal goat serum (S-1000; Vector Laboratories), 50% animal free blocker (SP-5035-100, Vector Laboratories) and 0.1% Triton X-100 in PBS) for 2 h, and then incubated with primary antibody diluted in blocking solution overnight at 4 °C. The sections were washed and then incubated with a secondary antibody diluted in a blocking solution for 4 h. After washing, the sections were counterstained with 4,6-diaminodino-2-phenylindole (DAPI, 1: 50,000; D1306; Thermo Fisher Scientific Inc.).
For immunoperoxidase labeling, tissue sections were washed, and incubated in a blocking solution mixed with Avidin D (1:50; Avidin/Biotin Blocking kit; SP-2001; Vector Laboratories Inc.) for 2 h, washed with PBS, then incubated in primary antibody diluted with blocking solution containing biotin (1:50; Avidin/Biotin Blocking kit) overnight at 4 °C. The sections were washed, incubated with biotinylated secondary antibody in blocking solution for 4 h, washed, incubated with 0.1% Triton X-100 in PBS containing avidin and biotin complex (1:50 each; Vectastain ABC Elite kit; PK-6100; Vector Laboratories Inc.) for 2 h, washed, then treated with 3,3′-diaminobenzidine (DAB) solution (DAB substrate kit; SK-4100; Vector Laboratories Inc.). Following labeling, the sections were washed and counterstained with DAPI. For ocular sections of newt larvae, melanin pigments of the RPE layer were bleached with PBS containing 3% H2O2 and 1.5% sodium azide26.
Gel shift assay
The WT (LexA binding element), TAx9, and GCx10 (negative control) containing single-strand oligonucleotides were self-annealed using conventional molecular cloning techniques to form duplex DNAs. In brief, the forward and reverse single-strand oligos of equal concentration (Supplementary Table 2) were heated at 95 °C for 3 min, incubated at 85 °C for 1 min, and then subjected to the same treatment until a total of 60 steps, where the annealing temperature was lowered by 1 °C in each successive step. Complexes of the E. coli LexA recombinant protein (01-005; Cosmo Bio, Tokyo, Japan) with each duplex DNA were formed according to established optimized methods for the WT (i.e., the SOS box) duplex DNA7. The protein-DNA reaction products (10 μL) were electrophoresed on 3.2% agarose gels (Agarose 21; 313-03242; Nippon Gene, Tokyo, Japan) in 0.5x TBE (45 mM Tris-borate, 1 mM EDTA) buffer at 4 °C and 100 V for 70 min.
Statistics and reproducibility
For E. coli, mice, and newts, at least three samples (experimental unit: a bacterial colony or a single animal) were analyzed per examination. No statistical comparisons of means or variances among multiple groups were made in this study. No data were excluded from the analyses. A minimum of three biological replicates were used in each evaluation since we employed only already established techniques in this study. In the case of recombination experiments, CreERT2–loxP animals were randomly separated into two groups before tamoxifen (test group) or DMSO/corn oil (control group) administration. There was no need for blind grouping because CreERT2–loxP animals with similar EGFP fluorescence were used. Data were evaluated independently by several authors. Potential confounders were not considered.
Microscopy and image analysis
Fluorescence of living newts and tail tips of mice were observed under a fluorescence dissecting microscope (M165 FC; Leica Microsystems, Tokyo, Japan) equipped with filter sets for EGFP (Leica GFP-Plant; exciter: 470/40 nm; emitter: 525/50 nm) and mCherry (exciter: XF1044, 575DF25; emitter: XF3402, 645OM75; Opto Science, Tokyo, Japan), which were designed to minimize bleed-through artifacts. Images were captured with a digital camera system (EOS Kiss x7i; Canon, Tokyo, Japan) attached to the microscope. Tissue sections were observed on a fluorescence microscope (BX50; Olympus, Tokyo, Japan) equipped with the same filter sets for EGFP and mCherry as well as that for DAPI. Images were captured with a charge-coupled device camera system (DP73; cellSens Standard 1.6; Olympus) attached to the BX50 microscope. For mouse limb sections, a laser confocal microscope system (LSM700; ZEN 2009, ver. 6.0.0.303; Carl Zeiss, Oberkochen, Germany) with filter sets for EGFP (Diode 488 Laser; emitter: BP 515–565 nm) and mCherry (Diode 555 Laser; emitter: BP 575–640 nm) was used. Images were analyzed with the software of the image acquisition systems and the functions of Adobe Photoshop 25.11.0 (Adobe, San Jose, CA, USA). The brightness, contrast, and sharpness of images were adjusted according to the journal’s guidelines using Photoshop 25.11.0. All gel electrophoresis data employed in this study are images of a single uncut gel containing size markers (for raw data, see Supplementary Fig. 11). Some images were inverted (Figs. 1–3 and Supplementary Fig. 5). Panels and Figures were prepared using Illustrator 27.4.1 graphics software (Adobe).
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
