Simple methods for generating and detecting locus-specific mutations induced with TALENs in the zebrafish genome

Abstract

The zebrafish is a powerful experimental system for uncovering gene function in vertebrate organisms. Nevertheless, studies in the zebrafish have been limited by the approaches available for eliminating gene function. Here we present simple and efficient methods for inducing, detecting, and recovering mutations at virtually any locus in the zebrafish. Briefly, double-strand DNA breaks are induced at a locus of interest by synthetic nucleases, called TALENs. Subsequent host repair of the DNA lesions leads to the generation of insertion and deletion mutations at the targeted locus. To detect the induced DNA sequence alterations at targeted loci, genomes are examined using High Resolution Melt Analysis, an efficient and sensitive method for detecting the presence of newly arising sequence polymorphisms. As the DNA binding specificity of a TALEN is determined by a custom designed array of DNA recognition modules, each of which interacts with a single target nucleotide, TALENs with very high target sequence specificities can be easily generated. Using freely accessible reagents and Web-based software, and a very simple cloning strategy, a TALEN that uniquely recognizes a specific pre-determined locus in the zebrafish genome can be generated within days. Here we develop and test the activity of four TALENs directed at different target genes. Using the experimental approach described here, every embryo injected with RNA encoding a TALEN will acquire targeted mutations. Multiple independently arising mutations are produced in each growing embryo, and up to 50% of the host genomes may acquire a targeted mutation. Upon reaching adulthood, approximately 90% of these animals transmit targeted mutations to their progeny. Results presented here indicate the TALENs are highly sequence-specific and produce minimal off-target effects. In all, it takes about two weeks to create a target-specific TALEN and generate growing embryos that harbor an array of germ line mutations at a pre-specified locus.

Figure 1. Induction of somatic golden mutations with TALENs.

(A) Schematic representation of the genomic structure of the golden (gol) gene, with coding and untranslated exon regions depicted as solid and open boxes, respectively. The locations of the gol-ex2 TALEN target site in exon 2 and the b1 null mutation in exon 5 are indicated. The gol-ex2 TALEN target sequence, with Left and Right TALEN monomer binding sites underlined, is shown. (B–F) Induction of golden mutant cells in the Retinal Pigmented Epithelium (RPE) of heterozygous gol^b1/+ embryos. Whereas the entire RPE of a 2 dpf heterozygous gol^b1/+ control embryo (B) is darkly pigmented, the RPEs of gol^b1/+ embryos injected with gol-ex2 TALEN RNAs (100 pg total) had patches of pigmentless mutant tissue (C–F). (G–M) Induction of golden mutant cells in the soma of wildtype gol^+/+ embryos. Wildtype gol^+/+ control embryos have darkly pigmented RPEs (G) and dark melanophores scattered over their bodies (L). Following injection at the 1 cell stage with 100 pg gol-ex2 TALEN RNAs, gol^+/+ embryos had patches of pigmentless golden mutant tissue in the RPEs (H–K) and some injected embryos appeared entirely devoid of pigmentation (M).

Figure 2. HRMA detects the presence of mutant alleles in mixed populations of wildtype and mutant genomes.

(A) Schematic illustration of the principle of HRMA. A small region of the genome that includes a DNA polymorphism is amplified by PCR from a mixed population of wildtype (WT) and mutant template genomes. Heat denaturation of the amplicons followed by rapid re-annealing of DNA strands produces two kinds of homoduplex and two kinds of heteroduplex products. Heteroduplex products are likely to be particularly unstable and exhibit a Tm that differs from the homoduplex products. (B) HRMA of lef1 PCR amplicons generated from template genomes that were either WT (lef1^+/+) or heterozygous for an 18 bp deletion (lef1^Δ18/+). PCR amplifications were performed in the presence of LC Green Plus dye (Idaho Technology), which fluoresces upon binding double strand DNA, allowing the detection of intact duplex molecules. The DNAs amplified from the lef1^+/+ genome (grey curves) comprise a homogeneous population of duplexes with a single Tm. In contrast, re-annealed amplicons derived from the lef1^Δ18/+ genome (red curve) are composed of multiple duplex populations, which display distinct Tms (*). (C) Sensitivity of HRMA for detecting the presence of the lef1^Δ18 mutation in mixtures of lef1^+/+ WT and lef1^Δ18/+ heterozygous genomes. Heterozygous DNA was mixed with increasing amounts of WT DNA; fractions indicate the relative abundance of mutant haploid genomes in the mixtures. A constant total amount of genomic template DNA and primers was used to generate each set of amplicons. LightScanner Call-IT Software (Idaho Technology) was used to identify melt curves that differed significantly from WT. HRMA detected the lef1 18 bp deletion allele when it was present as only 1/70^th of the total haploid genomes.

Figure 3. Induction of somatic mutations with TALENs.

(A–D, Upper panels) HRMA detection of targeted mutations in WT embryos that had been injected at the 1 cell stage with RNAs encoding TALENs directed against the golden (A), ryr3 (B), tbx6 (C), or ryr1a (D) gene (see Figure S1 for target sites). gDNA was isolated from individual uninjected or TALEN RNA-injected 1–2 dpf embryos and subjected to HRMA. Each curve is the melting profile of re-annealed amplicons generated from a single embryo. LightScanner Call-IT Software (Idaho Technology) was used to identify melt curves that differed significantly from WT. The results shown here reflect a single experiment, but results from all injections are tablulated in Table 2. Newly induced DNA polymorphisms at the targeted loci were evident in all but one injected embryo (red curves, Table 2), and even gol-ex2 TALEN RNA-injected embryos that did not exhibit patches of pigmentless tissue had induced golden mutations as detected by HRMA (A, purple curves). (A–C, Lower panels) TALEN-induced sequence alterations at targeted loci. Genomic sequences bordering the targeted loci were amplified from embryonic gDNA samples, cloned, and sequenced. Examples of recovered alleles (purple and pink boxed sequences indicate Left and Right RVD binding sites of the TALENs, respectively; red indicates sequence alterations) indicate that insertion/deletion (indel) mutations centered at the TALEN target sites had been induced in somatic tissues of embryos.

Figure 4. Dose-dependent induction of mutations with TALENs.

WT embryos were either not injected or injected at the 1 cell stage with 4 pg, 20 pg, or 100 pg total gol-ex2 TALEN RNA. gDNA was prepared from individual 1 dpf embryos and analyzed by HRMA for the presence of targeted mutations. Every embryo injected with 20 or 100 pg total RNA had mutations, but some 4 pg RNA-injected embryos did not have clear evidence of induced mutations (dark blue). As the amount of injected RNA was increased, the melt profiles of injected embryos, on average, diverged increasingly from the WT curves, indicating that increasing amounts of TALENs produce increasing numbers of mutant genomes.

Figure 5. Induction of germ line mutations with TALENs.

(A) TALEN RNA-injected G0 founders transmit targeted mutations to their offspring. G0 founders were mated with WT partners and the gDNAs of individual 1 dpf F1 progeny were examined by HRMA for the presence of mutations. The region flanking targeted sites was cloned and sequenced following PCR amplification from the gDNA of F1 offspring. Examples of germ line-transmitted mutations at each targeted locus are shown (purple and pink boxed sequences indicate Left and Right RVD binding sites of the TALENs, respectively; red indicates sequence alterations). (B, C) The germ lines of individual G0 founders are mosaic and can harbor several targeted mutations. Sibling F1 progeny of a single G0 founder targeted at gol (B) or a single G0 founder targeted at ryr3 (C) were examined by HRMA for inheritance of new mutations at gol or ryr3, respectively. Some F1 embryos (+/+) had no mutations, as their HRMA profiles (blue) resembled that of control WT embryos (grey). Other F1 embryos (m/+) displayed HRMA profiles indicating heterozygosity for a newly induced mutant allele (red). Several distinct HRMA profiles were identified among the heterozygous F1 progeny of a single G0 founder, indicating multiple mutations were transmitted by each G0 founder.

Figure 6. TALEN activity at homologous target sequences.

(A) Comparison of the ryr3 gene sequences targeted by the ryr3-ex5 TALEN and homologous sequences of the ryr1a and ryr1b genes. ryr3 sequences bound by the ryr3-ex5 TALEN are denoted in color, as in Figure S1: Red letters indicate the Left and Right RVD array binding sites of the TALEN monomer components; the entire target site is defined as including both the nucleotides recognized by RVD array and the 5′ T that interacts with the N-terminal portion of the TALEN. Vertical lines indicate bases that are identical between the ryr3-ex5 TALEN target and the gene sequences of homologues; *'s indicate non-identical bases. (B) Uninjected WT embryos (n = 23, grey curves) and WT embryos injected with 100 pg total ryr3-ex5 TALEN RNAs (n = 24, blue curves) were analyzed by HRMA at 1 dpf for the presence of newly induced mutations. The gDNA of each embryo was examined in parallel for mutations at ryr3, ryr1a, or ryr1b. Whereas every injected embryo had ryr3 mutations, no mutations were detected in the ryr1a or ryr1b genes. (C) Comparison of the ryr1a gene sequences targeted by the ryr1a-ex6 TALEN and homologous sequences of the ryr3 and ryr1b genes. ryr1a sequences bound by the ryr1a-ex6 TALEN are denoted in color, as in (A). Identical and non-identical bases are indicated as in (A). (D) Uninjected WT embryos (n = 23, grey curves) or WT embryos injected with 100 pg total ryr1a-ex6 TALEN RNAs (n = 24, red curves) were analyzed by HRMA at 1 dpf for the presence of newly induced mutations. Whereas no mutations at the ryr3 locus were detected in these embryos, every injected embryo had mutations in the ryr1a and ryr1b genes.