Protocol for generating splice isoform-specific mouse mutants using CRISPR-Cas9 and a minigene splicing reporter

Summary

Here, we present a protocol to alter the production of alternatively spliced mRNA variants, without affecting the overall gene expression, through CRISPR-Cas9-engineered genomic mutations in mice. We describe steps for designing guide RNA to direct Cas9 endonuclease to consensus splice sites, producing transgenic mice through pronuclear injection, and screening for desired mutations in cultured mammalian cells using a minigene splicing reporter. Splice isoform-specific mouse mutants provide valuable tools for genetic analyses beyond loss-of-function and transgenic alleles.

For complete details on the use and execution of this protocol, please refer to Dailey-Krempel et al.¹ and Johnson et al.²

Graphical abstract

Highlights

• •Introduction of genomic mutations in close proximity to a consensus splice site
• •Construction of a minigene splicing reporter for in vitro assays
• •Selection of desired splice isoform mutations using cultured mammalian cells

Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

Here, we present a protocol to alter the production of alternatively spliced mRNA variants, without affecting the overall gene expression, through CRISPR-Cas9-engineered genomic mutations in mice. We describe steps for designing guide RNA to direct Cas9 endonuclease to consensus splice sites, producing transgenic mice through pronuclear injection, and screening for desired mutations in cultured mammalian cells using a minigene splicing reporter. Splice isoform-specific mouse mutants provide valuable tools for genetic analyses beyond loss-of-function and transgenic alleles.

Before you begin

Through previous studies, we have demonstrated that alternative splicing of the Dcc (Deleted in Colorectal Carcinoma) gene is critical for its function and have identified two splice acceptor sites that produce the Dcc_Long/Dcc_L and Dcc_Short/Dcc_S isoforms, respectively (Figure 1A).^1,3,4 To generate isoform-specific mouse models, we used the protocol described here to introduce DNA mutations in close proximity to each of the two splice sites (Figure 1B), with the initial intention of disrupting and blocking their usage. Upon examining the effects of identified mutations using splicing assays in cultured mammalian cells, we found unexpectedly that the mutations in fact enhanced the usage of the targeted splice acceptor sites.¹ Our finding thus demonstrates that it is critical to employ complementary in vivo and in vitro assays as outlined in this protocol, to empirically determine how genomic DNA mutations may impact the outcome of alternative splicing.

Figure 1.

Design of guide RNAs (gRNAs) for Dcc splice isoforms

(A) Alternative splicing of 60 nt (in gray) at the 5′ end of exon 17 generates Dcc_L and Dcc_S isoforms.

(B) For each alternative splice acceptor site (in red letters), a gRNA (underlined 20 nt sequence) is designed to introduce double-strand break (DSB, indicated with scissors) in close proximity.

We describe here the steps to introduce Cas9-mediated DNA double-strand breaks (DSB) to two alternative splice acceptor sites at the intron 16/exon 17 boundary of the Dcc locus (Figure 1B). The subsequent DNA repair through the non-homologous end joining (NHEJ) mechanism introduces small insertions and/or deletions (indels).^5,6 To accurately predict how the indels may affect alternative splicing, we describe here the steps to construct a minigene splicing reporter, express the minigene in mammalian cells to produce alternative mRNA products, and to assess the mRNA species using reverse-transcription (RT) and semi-quantitative PCR. By comparing the splicing patterns produced from the wild-type and mutant reporters, we have selected two groups of mouse mutants where Dcc_L and Dcc_S are exclusively produced, respectively.¹

We have also used this protocol to design gRNAs for a homologous alternative exon (i.e., exon 6b) in the Robo1 and Robo2 loci, to construct a minigene reporter for exon 6b, and to identify mutations that exclusively produce exon6b-skipping isoforms of Robo1 and Robo2 (see Johnson et al.²).

Institutional permissions

All experimental manipulations and care of mice were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Colorado Boulder and the University of Minnesota Medical School. Users of this protocol should obtain IACUC approval from the relevant institutions before starting the experiments.

Design gRNAs to introduce indels in close proximity to isoform-specific splice sites

Timing: 1 day

The steps below describe the considerations in picking the top gRNA candidate to direct Cas9 nuclease to a consensus splice site.

• 1.
  For each target splice site, generate a candidate gRNA list using CRISPick. troubleshooting 1.

  Alternatives: Many additional gRNA design tools are available. For example, the web-based tool from IDT (https://www.idtdna.com/site/order/designtool/index/CRISPR_CUSTOM) can be used for both designing and ordering gRNAs. Users may also consider using different design tools to select a consensus gRNA target.

  • a.
    Go to https://portals.broadinstitute.org/gppx/crispick/public.

    • i.
      Under Reference Genome, select either Mouse GRCm38 (NCBI RefSeq) or Mouse GRCm38 (Ensembl). The two genome annotation methods do not affect the outcome of candidate selection.
    • ii.
      Under Mechanism, select CRISPRko.
    • iii.
      Under Enzyme, select SpyoCas9 and then Hsu (2013) tracrRNA.
    • iv.
      Under Target(s), select Bulk/Advanced targets and enter the sequence from the target genomic region.

      Note: We recommend entering a short sequence (50–100 nt) that centers around the desired DNA break location (e.g., at the splice acceptor site for Dcc_L or Dcc_S). The entered sequences for Dcc isoforms are as follows: Dcc_L 5′-TGATATTTATTTTGACTCTTCGTTTACACATTTATTTGTTTGTTTCATTTGGTGGGTTTTGAACCCAGATCCCACTGACCCCGTTGATTATTATCCTTTG-3′ and Dcc_S 5′-GTTGATTATTATCCTTTGCTTGATGATTTCCCCACCTCGGGCCCAGATGTCTCCACCCCCATGCTCCCACCAGTAGG-3’.

    • v.
      Click Validate to submit the sequence.
  • b.
    On the next page of input validation, click Submit.
  • c.
    On the next page, click Picking Results to download the file containing gRNA candidates. Open the file with Excel.
• 2.
  Select the top gRNA(s) from the returned candidates using the following criteria.

  • a.
    The proximity between the expected DSB and the target splice site (generally, the closer, the better).

    Note:S. pyogenes Cas9 (SpCas9) typically introduces DSB 3–4 bp upstream of the NGG PAM (protospacer adjacent motif) sequence [the PAM sequence is immediately 3′ to the 20 nt gRNA sequence and is required for Cas9 to introduce DSB].⁷

  • b.
    The location of the DSB in intron vs. exon sequences (generally, cutting in an intron is preferred).

    Note: For example, gRNA #2 for Dcc_L splice site will likely introduce a DSB and subsequent indel mutations in intron 16, but not in exon 17 (Figure 2A). This gRNA is thus unlikely to disrupt the translation of DCC_L protein.

  • c.
    The overall ranking based on the balance between on-target and off-target scores (generally, the higher, the better).

    Note: When more than one gRNA is needed to introduce the desired mutations, we recommend selecting candidates with comparable on-target scores so that DSBs can be introduced at comparable probabilities.

    Note: Taking everything into account, our top candidates for Dcc_L and Dcc_S are gRNA #2 and #3, respectively (Figure 2). The expected DSBs are within 20 nt from the respective splice sites and are located within an intron (for the Dcc_S splice site, the alternative 60 nt of exon 17 is considered intronic sequence).

• 3.
  Purchase the selected single guide RNA (sgRNA) from a commercial vendor [We use Integrated DNA Technology (IDT)].

  Note: sgRNA is an RNA molecule that contains both CRISPR RNA (crRNA), the 20 nt sequence selected in step 2 to recognize target genomic DNA, and tracrRNA, the binding scaffold for Cas9 nuclease.⁸ When designing gRNAs for a large number of targets, users may consider ordering two-part gRNAs, with crRNA and tracrRNA present in separate RNA molecules. While crRNA is unique for each target sequence, tracrRNA is common.⁸

  • a.
    Go to CRISPR-Cas9 gRNA checker webpage at IDT https://www.idtdna.com/site/order/designtool/index/CRISPR_SEQUENCE.
  • b.
    Select the species (i.e., mouse) and enter the selected gRNA sequence from step 2 (do not include the NGG PAM sequence at the 3′ end). Click CHECK.
  • c.
    On the next page, which displays the on-target and off-target scores based on IDT’s algorithm, click Quick Order and then select the scale and format of the product.
  • d.
    After receiving the sgRNA, dissolve it in nuclease-free water with 10 mM Tris-HCl, pH 8.0, at a stock concentration of 20 μM.

    Pause point: Store the sgRNAs at −80°C until ready for pronuclear injection.

Figure 2.

Comparison of top gRNA candidates

(A and B) Top five gRNA candidates for Dcc_L and Dcc_S splice sites (in red letters), respectively. Lines with an arrow indicate the target sequence (20 nt) and the 5′ to 3′ orientation. Numbers under each line indicate the overall ranking. The top gRNA for each isoform is shown in dark red, and the unselected gRNAs in pink. The DSB introduced by Dcc_S gRNA is likely to eliminate an ApaI digestion site (in blue letters).

Molecular cloning and in vitro transcription of Cas9 nuclease

Timing: 2 weeks

The steps below use in vitro transcription from a T7 promoter to generate Cas9 mRNA, which will be injected together with sgRNA into mouse zygotes to produce transgenic mice.

Alternatives:Cas9 mRNA is also commercially available (e.g., from IDT). If Cas9 protein (commercially available) will be injected together with gRNA or if Cas9- and gRNA-encoding DNA will be injected, skip this step.

• 4.
  Subclone Cas9 coding sequence downstream of a T7 promoter.

  Alternatives: Alternative promoters, e.g., T3 or SP6, can be used for in vitro transcription. Users should choose the subcloning vector and in vitro transcription reagents accordingly.

  • a.
    Obtain the pX330 plasmid from Addgene (#42230), which contains a humanized coding sequence of SpCas9.⁹
  • b.
    Design PCR primers to amplify the complete Cas9 coding sequence.

    Note: To enhance protein translation, include a Kozak consensus sequence before the start codon. There is no need to include restriction sites in the primers.

    CRITICAL: Besides the Cas9 start codon included in the forward primer, a second ATG sequence is present in the cloning vector downstream of the T7 promoter. Make sure that the two ATG sequences are in the same reading frame. Otherwise, a different peptide may be translated from the transcript.

  • c.
    Perform PCR to amplify Cas9 coding sequence from pX330.

    Note: Use a high-fidelity DNA polymerase (e.g., we use Phusion polymerase) to avoid introducing mutations. Assemble and perform the PCR reaction following the manufacturer’s protocols.

  • d.
    Add 6x gel loading dye to the reaction and perform DNA electrophoresis in a 1% agarose gel in 1x TAE buffer (at 140 V for ∼30 min). Include a lane with 1 kb DNA Ladder.
  • e.
    Using the ChemiDoc Imaging System, excise the band of the correct size from the agarose gel with a blade (use 450–490 nm ultraviolet wavelength to minimize damage to DNA molecules).
  • f.
    Purify the DNA using QIAquick Gel Extraction Kit.
  • g.
    Clone the purified amplicon into the pJET1.2/blunt vector using CloneJET PCR cloning kit, following the manufacturer’s protocol.
  • h.
    Transform 1–3 μL of the reaction into chemically competent E. coli (e.g., DH5α). After heat shock at 42°C for 30 s, add 250 μL of SOC media and allow the bacteria to recover at 37°C with shaking for 30–60 min.
  • i.
    Spread the bacteria on an LB agar plate containing 100 μg/mL ampicillin, and incubate for ∼16 h at 37°C.
  • j.
    The next day, pick a few single colonies to start a 5 mL culture of LB broth containing 100 μg/mL ampicillin. Incubate for ∼16 h at 37°C with vigorous shaking.
  • k.
    Purify the plasmid DNA using QIAprep Spin Miniprep Kit.
  • l.
    Perform restriction digestion to confirm that Cas9 coding sequence is inserted in the right orientation (i.e., the sense strand of Cas9 will be transcribed from the T7 promoter). For example, NotI and SacI digestion should produce 2.4 and 4.8 kb fragments.
  • m.
    Verify the sequence using whole plasmid sequencing (e.g., we use Eurofins Genomics).
• 5.
  Generate Cas9 mRNA through in vitro transcription.

  CRITICAL: Use RNase-free reagents and plasticware for steps c-h to avoid degradation of in vitro transcribed mRNA.

  • a.
    Linearize the Cas9/pJET1.2 plasmid from step 4 with XbaI, which is located downstream of Cas9 stop codon. Assemble the restriction digestion reaction at 20°C–25°C (see materials and equipment) and incubate at 37°C for 1 h.
  • b.
    Load the digested DNA in a 1% agarose gel and run electrophoresis.
  • c.
    Excise the linearized DNA band and purify the DNA using QIAquick Gel Extraction Kit.
  • d.
    Using the reagents from mMESSAGE mMACHINE T7 ULTRA Transcription Kit, assemble the in vitro transcription reaction at 20°C–25°C (see materials and equipment). Incubate at 37°C for 2 h.
  • e.
    Add 1 μL of TURBO DNase provided in the kit and incubate at 37°C for 15 min to remove the DNA template.
  • f.
    Add the tailing reagents in the kit, and incubate at 37°C for 45 min.

    Note: The final mRNA product contains a 5′ cap analog and a 3′ poly(A) tail, which enhances the stability and translation efficiency of the mRNA.

  • g.
    Purify the mRNA product using MEGAclear Transcription Clean-Up Kit.

    • i.
      Bring the volume of the in vitro transcribed RNA to 100 μL with Elution Solution in the kit.
    • ii.
      Add 350 μL of Binding Solution Concentrate to the sample. Mix by pipetting.
    • iii.
      Add 250 μL of 100% ethanol to the sample. Mix by pipetting.
    • iv.
      Transfer all solution into a Filter Cartridge inserted into a collection tube. Spin the Filter Cartridge at 10,000 × g for 1 min and discard the flowthrough.
    • v.
      Add 500 μL of Washing Solution (with ethanol added) to the Filter Cartridge and spin at 10,000 × g for 1 min. Discard the flowthrough.
    • vi.
      Repeat step v. to wash the Filter Cartridge a second time.
    • vii.
      Dry the Filter Cartridge by spinning the tube at 10,000 × g for 30 s.
    • viii.
      Transfer the Filter Cartridge to a clean collection tube. Add 50 μL of Elution Solution to the Filter Cartridge, heat at 70°C for 10 min.
    • ix.
      Centrifuge the tube at 10,000 × g for 1 min to elute the mRNA.
  • h.
    Measure the mRNA concentration using a NanoDrop spectrophotometer, and examine the size of the mRNA using gel electrophoresis.

    Alternatives: Assess the mRNA size and quality using a Bioanalyzer (Agilent).

    • i.
      To 500 ng-1 μg RNA, add an equal volume of 2x RNA loading buffer.
    • ii.
      Load the mix in a 1% agarose gel in 1xTAE buffer. Include one lane of RiboRuler high range RNA Ladder for size comparison.
    • iii.
      Run gel electrophoresis at 140 V for 20–30 min.
    • iv.
      Image the RNA gel using the ChemiDoc Imaging System.

      Pause point: Store Cas9 mRNA at −80°C until ready for pronuclear injection.

Construct a minigene splicing reporter to assess alternative splicing in vitro

Timing: 2–3 weeks

A minigene splicing reporter contains a selected segment of genomic DNA encompassing the alternatively spliced sequence, and is routinely used to study the splicing events of interest independent of other regions in the pre-mRNA.¹⁰ The steps below produce a splicing reporter for Dcc genomic DNA between exons 16 and 17, using PCR amplification and DNA subcloning.

Alternatives: Users may select their preferred cloning method and expression vector. For example, TOPO cloning can be used to directly insert a PCR product into an expression vector (e.g., pcDNA3.2/V5/GW/D-TOPO, which contains a CMV promoter to drive gene expression). Other recombination-based cloning method, such as NEBuilder HiFi DNA Assembly, can be used to clone PCR-amplified genomic DNA into any vector of choice.

• 6.
  Design PCR primers to amply the genomic segment to be included in the minigene reporter.

  • a.
    Identify the 5′ and 3′ flanking exons of the alternatively spliced sequence in the mature mRNA, i.e., exons 16 and 17 of Dcc (see Figure 1).
  • b.
    Design forward and reverse PCR primers that start from the very end of the 5′ and 3′ flanking exons, respectively. Add attB1 sequence to the 5′ end of the forward primer, and attB2 sequence to the reverse primer to enable Gateway cloning (see key resources table for primer sequences).

    Note: There is no need to include a start or stop codon in the primers because the minigene reporter is assessed only at the mRNA level and the vector provides stop codons and a poly adenylation signal.

  • c.
    Have the primers synthesized (e.g., at IDT).
• 7.
  Isolate mouse genomic DNA using phenol/chloroform extraction.

  Alternatives: Ready-to-use mouse genomic DNA as well as column-based genomic DNA extraction kits are commercially available (e.g., DNeasy Blood & Tissue Kits from QIAGEN).

  • a.
    Collect a piece of mouse tissue (e.g., a tail clip) into a 1.5 mL centrifuge tube.
  • b.
    Add 200 μL of tissue lysis buffer containing Proteinase K (see materials and equipment) to the tube, and incubate at 50°C–60°C for ∼16 h with shaking.
  • c.
    Add 200 μL of phenol/chloroform/isoamyl alcohol (25:24:1). Mix by inverting the tubes vigorously. Let sit at 20°C–25°C for 10 min.
  • d.
    Spin at >10,000 × g for 5 min in a benchtop centrifuge and transfer the top aqueous phase (∼200 μL) to a fresh tube.
  • e.
    To precipitate the DNA, add 20 μL of 3 M sodium acetate (or 1/10 volume) and 160 μL of isopropanol (0.7 volume) to the supernatant. Vortex to mix and incubate at 20°C–25°C for 10–15 min.
  • f.
    Centrifuge at >10,000 × g for 10 min and discard the supernatant.
  • g.
    Add 250 μL of 70% ethanol to the tube and centrifuge at >10,000 × g for 5 min.
  • h.
    Carefully aspirate the supernatant and air dry the pellet for 5–10 min.
  • i.
    Add 50 μL of elution buffer (10 mM Tris, pH 7.5) to resuspend the DNA. Vortex and incubate at 50°C–60°C for 10 min.
• 8.
  Using primers designed in step 6, perform PCR to amplify genomic DNA encompassing the alternatively spliced region. troubleshooting 2. troubleshooting 3.

  • a.
    Set up and run the PCR reaction using Phusion polymerase following the manufacturer’s protocol.
  • b.
    Examine the PCR product using gel electrophoresis.
  • c.
    Excise the correct band, and purify the DNA using QIAquick Gel Extraction Kit.
• 9.
  Clone the purified PCR product into an expression vector [we use pDEST26 and the Gateway cloning method with two-step recombination reactions (see Figure 3)].

  Alternatives: To save time, the cloning process can be reduced to a one-step reaction by mixing the PCR product, pDONR221 donor vector, pDEST26 expression vector, and LR clonase II enzyme mix in a single tube.¹¹

  • a.
    Set up the first recombination reaction (BP reaction) using the PCR product and the pDONR221 donor vector (see materials and equipment). Incubate at 25°C for 1 h.
  • b.
    Add 1 μL Protease K solution to the reaction and incubate at 37°C for 10 min.
  • c.
    Transform 1 μL of the reaction into chemically competent E. coli (e.g., DH5α), spread the bacteria on a LB agar plate containing 50 μg/mL kanamycin, and incubate at 37°C for ∼16 h until the next day.
  • d.
    Select a few colonies to start a 5 mL liquid culture of LB with kanamycin and incubate for ∼16 h at 37°C with vigorous shaking.
  • e.
    Purify the resulting entry clone plasmid using QIAprep Spin Miniprep Kit.
  • f.
    Validate the entry clone with restriction digestion (e.g., EcoRV digestion produces 4.5 kb and 6.9 kb fragments).
  • g.
    Set up a second recombination reaction (LR reaction) using the verified entry clone and the pDEST26 destination vector (see materials and equipment). Incubate at 25°C for 1 h.
  • h.
    Add 1 μL Protease K solution to the reaction and incubate at 37°C for 10 min.
  • i.
    Transform 1 μL of the reaction into chemically competent E. coli (e.g., DH5α), spread the bacteria on a LB agar plate containing 100 μg/mL ampicillin, and incubate for ∼16 h at 37°C.
  • j.
    Select a few colonies to start a 5 mL liquid culture of LB with ampicillin and grow for ∼16 h at 37°C with vigorous shaking.
  • k.
    Purify the resulting expression clone, i.e., the minigene splicing reporter, using QIAprep Spin Miniprep Kit.
  • l.
    Validate the minigene reporter with restriction digestion (e.g., SacI and EcoRV digestion produces 1.4 kb and 10 kb fragments).
  • m.
    Verify the minigene construct by whole plasmid sequencing.

Figure 3.

Two-step Gateway cloning process

Perform splicing assays in cultured mammalian cells

Timing: 1 week

The steps below describe transfection of the minigene reporter into cultured mammalian cells and detection of alternatively spliced mRNA products using reverse transcription and semi-quantitative PCR.

• 10.Subculture COS-1 cells and plate the cells in a 24-well culture dish in complete growth medium (see materials and equipment; use 500 μL medium per well). The cells are recommended to be at ∼80% confluency the next day.

Alternatives: Other easily transfectable cell lines, e.g., HEK293T and HeLa, also work well.

• 11.
  The next day, transfect the cells with the minigene reporter using TransIT-LT1 reagent.

  Alternatives: Other liposome-base reagents, e.g., Lipofectamine, also work well.

  • a.
    Aliquot 500 ng of minigene reporter (for each well) in a centrifuge tube.

    Optional: NOVA1 and NOVA2 splicing factors bind specific sequences in intron 16 of Dcc pre-mRNA and promote the usage of Dcc_L splice site.^3,12 We include additional conditions where NOVA1 or NOVA2 is overexpressed, to determine if the splicing pattern of the reporter recapitulates that in vivo. We use 250 ng of the minigene and 250 ng of pCAG empty vector³ for the negative control condition, and 250 ng of the minigene and 250 ng of pCAG-Nova1 or pCAG-Nova2 for the overexpression condition.³

  • b.
    Add 50 μL of serum free Opti-MEM to the DNA, and then pipet 1 μL of TransitLT1 directly to the solution. Vortex to mix and let sit at 20°C–25°C for 15 min.
  • c.
    Add the transfection mix to the cells.
• 12.
  24–48 h post-transfection, aspirate the growth medium and add direct lysis buffer to extract total RNA (modified from Ho et al.¹³).

  Alternatives: Total RNA can be extracted and purified using other reagents such as QIAzol or RNeasy Mini Kit (QIAGEN).

  CRITICAL: Use RNase-free reagents and plasticware to prevent RNA degradation.

  • a.
    Add 100–150 μL direct lysis buffer (see materials and equipment) to each well in a 24 well plate.
  • b.
    Incubate the plate at 20°C–25°C for 10 min with shaking.
  • c.
    Transfer the lysate to a centrifuge tube and heat at 75°C for 5 min to inactivate the DNase.
  • d.
    Centrifuge at >10,000 × g for 5 min at 4°C to pellet cells and debris.
  • e.
    Transfer the supernatant to a clean centrifuge tube.

    Pause point: Store total RNA at −80°C or proceed with reverse transcription.

• 13.Set up reverse transcription reactions to synthesize cDNA [we use SMARTScribe reverse transcriptase; see materials and equipment].

Note: We recommend using a thermocycler for all incubation steps.

Alternatives: Other first-strand cDNA synthesis reagents, e.g., SuperScript III, also work well.

Note: mRNA expressed from the pDEST26 vector contains a T7 sequence at the 3′ end in a reverse orientation. Thus, T7 primer is used to amplify Dcc mRNA transcribed from the minigene, but not from the endogenous locus in COS-1 cells.

• 14.
  Determine the relative expression of alternative isoforms using semi-quantitative PCR.

  • a.
    Set up a quantitative PCR (qPCR) using a primer pair that amplifies both splice isoforms (i.e., Dcc_L and Dcc_S). We use Terra qPCR Direct TB Green Premix and CFX Connect Real-Time PCR Detection System.

    Alternatives: Users may select their preferred DNA polymerase, DNA dye, and real-time thermocycler for qPCR.

  • b.
    Determine the cycle numbers that are within the exponential phase of amplification (Figure 4A).
  • c.
    Using the same primers as for qPCR and a cycle number within the exponential phase (e.g., 30), perform a regular PCR, which is a semi-quantitative reaction, to amply both isoforms.

    Note: We recommend using the same reaction mix as in qPCR to ensure the consistency in the amplification dynamics. We also recommend using the highest cycle number within the exponential phase to facilitate product detection.

  • d.
    Separate the amplicons using gel electrophoresis. We use 3% agarose gel to separate 241 and 181 bp amplicons for Dcc_L and Dcc_S, respectively.
  • e.
    Image the gel (Figure 4B; the wild-type splicing pattern will be compared to the mutant ones later in the protocol). troubleshooting 4.

    Optional: Quantify the ratio between the splice variants using FIJI, if needed.

Figure 4.

Semi-quantitative PCR to detect Dcc isoforms in splicing assays

(A) The amplification curve of Dcc isoforms in a quantitative PCR assay. Cycle numbers between 25 and 30 (indicated with dashed lines) are within the exponential phase of amplification. A cycle number of 30 is selected for semi-quantitative PCR to compare Dcc isoform expression.

(B) Alternative splicing patterns produced from wild-type and mutant minigenes in COS-1 cells. Two products are detected from wild-type minigene and the NOVA splicing factors promote Dcc_L production as in vivo. When 9 bp are deleted immediately upstream of Dcc_L splicing site, Dcc_L is constitutively expressed with or without the presence of NOVA.

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Bacterial and virus strains
Subcloning efficiency DH5α competent cells	New England Biolabs	C2987H
One Shot TOP10 chemically competent E. coli	Invitrogen	C404003
Chemicals, peptides, and recombinant proteins
PMSG (gonadotropin from pregnant mare serum)	MilliporeSigma	G4877
HCG (human chorionic gonadotropin)	MilliporeSigma	CG10
Hyaluronidase	MilliporeSigma	H3884
M2 medium	MilliporeSigma	M7167
M16 medium	MilliporeSigma	M7292
Mineral oil	MilliporeSigma	330779
DMEM medium	Corning	10-013-CV
Opti-MEM I reduced serum medium	Gibco	31985070
Fetal bovine serum (FBS)	Corning	35-010-CV
Penicillin-streptomycin	Corning	30-002-CI
Phusion DNA polymerase	New England Biolabs	M0530S
XbaI restriction enzyme	New England Biolabs	R0145L
ApaI restriction enzyme	New England Biolabs	R0114L
PacI restriction enzyme	New England Biolabs	R0547L
EcoRI-HF restriction enzyme	New England Biolabs	R3101L
RiboRuler high range RNA ladder	Thermo Fisher Scientific	SM 1821
1 kb DNA ladder	New England Biolabs	N3232L
100 bp DNA ladder	New England Biolabs	N3231L
Gel loading dye, purple (6X)	New England Biolabs	B7025S
RNA loading dye (2X)	New England Biolabs	B0363S
SOC outgrowth medium	New England Biolabs	B9020S
10x TAE buffer	Fisher Scientific	BP13354
TransIT-LT1 reagent	Mirus Bio	MIR 2300
Phenol/chloroform/isoamyl alcohol (25:24:1)	Invitrogen	15593031
Phenol/chloroform/isoamyl alcohol is combustible and is toxic when inhaled, consumed, or in contact with skin. Wear personal protective equipment when handling. Store and dispose of properly.
Critical commercial assays
mMESSAGE mMACHINE T7 ULTRA transcription kit	Invitrogen	AM1345
MEGAclear transcription clean-up kit	Invitrogen	AM1908
QIAprep spin miniprep kit	QIAGEN	27016
QIAquick gel extraction kit	QIAGEN	28706
Terra PCR direct red dye premix	Takara Bio	639286
Terra qPCR direct TB green premix	Takara Bio	638319
CloneJET PCR cloning kit	Thermo Fisher Scientific	K1231
TOPO TA cloning kit for subcloning	Invitrogen	450641
Gateway LR Clonase II enzyme mix	Invitrogen	11791020
Gateway BP Clonase II enzyme mix	Invitrogen	11789020
SMARTScribe reverse transcriptase	Takara Bio	639537
Experimental models: Cell lines
COS-1	ATCC	CRL-1650
Experimental models: Organisms/strains
Female adult FVB mice (for pronuclear injection)	The Jackson Laboratory	Strain #:001800 RRID:IMSR_JAX:001800
Female adult CD-1 mice (for embryo transfer)	Charles River Laboratories	Crl:CD1(ICR)
Male and female adult C57BL/6J mice (for outcross and backcross breeding)	The Jackson Laboratory	Strain #:000664 RRID:IMSR_JAX:000664
Oligonucleotides
Guide RNA & PAM for DccL splice site: ATTTGTTTGTTTCATTTGGTggg (PAM sequence in lower case)	Dailey-Krempel et al.1	N/A
Guide RNA & PAM for DccS splice site: GGTGGAGACATCTGGGCCCGagg (PAM sequence in lower case)	Dailey-Krempel et al.1	N/A
Cas9 forward primer: GCCACCATGGACTATAAGG ACCACGACGGA (this primer contains a Kozak sequence, a start codon, and a FLAG tag sequence, all of which are upstream of Cas9 coding sequence in pX330)	This paper	N/A
Cas9 reverse primer: TTACTTTTTCTTTTTTGCCTGG	This paper	N/A
Forward primer for amplifying Dcc minigene: acaagtttgtacaaaaaagcaggctAGTCGAGTTCTC ATTATGTAATCTCCTTA (attB1 in lower case)	This paper	N/A
Revere primer for amplifying Dcc minigene: accactttgtacaagaaagctgggtCTTGTACTTGG CACTGGCAGAAAA (attB2 in lower case)	This paper	N/A
Forward primer for quantitative and semi-quantitative PCR: TCTCATTATGTAATCTCCTTAAAAGC	Leggere et al.3	N/A
Reverse primer for quantitative and semi-quantitative PCR: TCACAGCCTCATGGGTAAGAG	Leggere et al.3	N/A
Forward primer for amplifying mutant sequence from F0 pups: AAGTTGGAGCCTTAAGGTATATATTC	This paper	N/A
Reverse primer for amplifying mutant sequence from F0 pups: TCACAGCCTCATGGGTAAGAG	This paper	N/A
Recombinant DNA
pX330 (contains humanized coding sequence of S. pyogenes Cas9)	Cong et al.9	Addgene 42230
SpCas9 in pJET1.2	This paper	N/A
Dcc minigene reporter	Leggere et al.3	N/A
pCR2.1TOPO	Invitrogen	K450002
Gateway pDONR221 vector	Invitrogen	12536017
Gateway pDEST26 vector	Invitrogen	11809019
pCAG	Leggere et al.3	N/A
pCAG-Nova1	Leggere et al.3	N/A
pCAG Nova2	Leggere et al.3	N/A
Software and algorithms
CRISPick (for guide RNA design)	Broad Institute, MIT	https://portals.broadinstitute.org/gppx/crispick/public
CRISPR-Cas9 gRNA Checker (for guide RNA design/synthesis)	IDT	https://www.idtdna.com/site/order/designtool/index/CRISPR_SEQUENCE
DNASTAR	DNASTAR, Inc.	N/A
Other
CFX Connect real-time PCR detection system	Bio-Rad	1855201
Eppendorf Mastercycler	Eppendorf	Model proS
ChemiDoc Imaging System	Bio-Rad	12003153
NARISHIGE IM 300 microinjector	Narishige International USA, Inc.	IM-300
Stereo microscope	Zeiss	Stemi SV 11 Apo
Light source for stereo microscope	Zeiss	KL 1500 LCD
Inverted Nomarski DIC microscope	Leica Microsystems	DMIRB
Micromanipulator	Leica Microsystems	N/A
Steri-Cycle CO2 incubator	Thermo Fisher Scientific	Model 370
NanoDrop One spectrophotometer	Invitrogen	D-ONE-W
Micropipette puller	Sutter Instrument	P-2000
Micropipettes, borosilicate glass with filament, O.D. 1.0 mm, I.D. 0.5 mm, 10 cm length	Sutter Instrument	BF 100-50-10
Owl EasyCast B1A mini gel electrophoresis systems	Thermo Fisher Scientific	B1A-BP

Materials and equipment

Restriction digestion

Component	Amount
Plasmid DNA	1–2 μg
10X CutSmart buffer	2 μL
Restriction enzyme(s)	0.5 μL each
ddH2O	To 20 μL

In vitro transcription of Cas9 mRNA

Component	Amount
linearized DNA template	1 μg
T7 2X NTP/ARCA	10 μL
10X T7 Reaction Buffer	2 μL
T7 Enzyme Mix	2 μL
Nuclease-free H2O	To 20 μL

Tissue lysis buffer for DNA extraction

Component	Final concentration	Amount
1 M Tris pH 8.0	10 mM	0.5 mL
5 M NaCl	100 mM	1 mL
0.5 M EDTA pH 8.0	10 mM	1 mL
10% SDS	0.5%	2.5 mL
ddH20	–	To 50 mL

Store at 20°C–25°C for up to a year. Add 20 μL of Proteinase K (20 mg/mL stock) per 1 mL of tissue lysis buffer immediately before use.

Reagent	Final concentration	Amount
10x RQ1 DNase reaction buffer	40 mM Tris pH 8, 10 mM MgSO4, 1 mM CaCl2	1 mL
RQ1 DNase, 1 U/μL	0.1 U/μL	1 mL
RNAsecure (25x)	1x	0.4 mL
Triton X-100, 20%	2%	1 mL
NP40, 20%	2%	1 mL
RNase & DNase free H2O	–	To 10 mL

Aliquot and store at −20°C for up to 6 months.

Direct lysis buffer for RNA extraction from cultured cells.

Gateway cloning reactions

BP reaction	Amount
PCR product (flanked by attB1 and attB2 sequences)	150 ng
pDONR221, 150 ng/μL	1 μL
TE buffer	To 8 μL
BP clonase II enzyme mix	2 μL
Total	10 μL
LR reaction	Amount
Entry clone	150 ng
pDEST26, 150 ng/μL	1 μL
TE buffer	To 8 μL
LR clonase II enzyme mix	2 μL
Total	10 μL

Reverse transcription of Dcc minigene reporter

Component	Final concentration	Amount
Direct RNA lysate	–	4 μL
Reverse primer T7 (20 μM)	2 μM	1 μL

Incubate at 72°C for 3 min, and immediately cool on ice.

Proceed to add the following components to the reaction.

5x First-strand buffer	1x	2 μL
dNTP mix, 10 mM each	1 mM each	1 μL
DTT (20 mM)	2 mM	1 μL
SMARTScribe RT enzyme	10 U/μL	1 μL
Total	–	10 μL

Incubate at 42°C for 60 min, and then inactivate the reaction at 70°C for 15 min.

Quantitative PCR for Dcc isoforms

Reagent	Final concentration	Amount
Reverse transcription reaction	–	2 μL
2 x Terra qPCR Direct TB Green Premix	1x	10 μL
Forward primer (10 μM)	0.5 μM	1 μL
Reverse primer (10 μM)	0.5 μM	1 μL
ddH2O	–	To 20 μL

See key resources table for primer sequences.

Step	Temperature	Time	Cycle
Initial denaturation	95°C	2 min	1
Denaturation	95°C	15 s	40
Annealing	60°C	15 s
Extension+ plate read	72°C	15 s (1 min/kb)
Denaturation	95°C	10 s	1
Melt curve+ plate read	65°C–95°C	5 s/step, 0.5°C increment/cycle

complete growth medium for COS-1 cells

Reagent	Final concentration	Amount
DMEM	1x	440 mL
Fetal bovine serum	10%	50 mL
Penicillin-streptomycin 50x solution	100 U/mL penicillin, 100 μg/mL streptomycin	10 mL

Store at 4°C for up to 4 weeks.

PCR using crude genomic DNA lysate

Reagent	Final concentration	Amount
Crude tissue lysate	–	2 μL
2 x Terra PCR Direct Red Dye Premix	1x	10 μL
Forward primer (10 μM)	0.5 μM	1 μL
Reverse primer (10 μM)	0.5 μM	1 μL
ddH2O	–	To 20 μL

Note: Glycerol and a red dye are included in the Premix solution, so the PCR products can be directly loaded into an agarose gel.

Step	Temperature	Time	Cycle
Initial denaturation	95°C	2 min	1
Denaturation	95°C	15 s	30
Annealing	60°C	15 s
Extension	72°C	20 s (1 min/kb)
Hold	4°C	indefinitely

Restriction digestion of unpurified PCR product

Reagent	Final concentration	Amount
PCR reaction	–	20 μL
10x CutSmart buffer	1x	4 μL
ApaI 50 U/μL	0.625 U/μL	0.5 μL
ddH2O	–	To 40 μL

Incubate at 37°C for 30 min.

HiFi DNA assembly reaction

Reagent	Amount
Linearized minigene vector (11 kb)	100 ng
Identified mutant genomic DNA (0.4 kb)	5 ng
2x NEBuilder HiFi DNA Assembly Master Mix	5 μL
ddH2O	To 10 μL

Incubate at 50°C for 15 min.

Step-by-step method details

Note: Before proceeding with pronuclear injection in the next step, readers may consider validating the candidate sgRNAs in cultured cells. While planning the cell culture experiment, bear in mind that individual cells will introduce independent and thus distinct repairs after DSB. In order to determine the effects of individual mutations, one will need technical tools to differentiate the isoforms at single-cell resolution. For example, a cell line that endogenously expresses the alternative isoforms will be needed, instead of using cells transfected with a minigene reporter. This is because cells are under survival pressure to repair the genomic DNA but not the exogenously transfected DNA. In addition, tools such as isoform-specific antisense probes or antibodies will be needed to compare the isoform expression in individual cells using in situ hybridization or immunocytochemistry. Because we did not have the necessary tools to detect Dcc isoforms in individual cells (although we found that Neuro2a cells endogenously express both Dcc isoforms), we proceeded with in vivo mutagenesis without validating the sgRNAs in vitro.

Produce founder (F0) transgenic mice through pronuclear injection of Cas9 mRNA and sgRNA

Timing: 3–4 weeks

The steps below describe collecting fertilized eggs (zygotes) at the pronuclear stage (when the maternal and paternal pronuclei have not yet fused into one), injecting Cas9 mRNA and sgRNA into the zygotes, and transferring the injected zygotes into the oviducts of surrogate female mice.

• 1.
  Superovulation and mating of egg donor females.

  Note: We use females of the FVB/NJ strain as egg donors due to the reported high efficiency of transgenic mouse production.¹⁴

  • a.
    Three days prior to pronuclear injection, administer PMSG (gonadotropin from pregnant mare serum, 5 IU per mouse) by intraperitoneal (i.p.) injection.

    Note: We typically prepare 10 donor females for each injection.

  • b.
    48 h later, administer HCG (human chorionic gonadotropin, 5 IU per mouse) by i.p. injection.
  • c.
    On the same day, set up mating pairs, each consisting of one superovulated female and one stud male.
  • d.
    Check for the presence of a vaginal plug the next morning.
• 2.
  Collect fertilized eggs at the pronuclear stage (procedure adapted from Ittner and Götz¹⁵).

  • a.
    Prepare the following solutions in 35 mm dishes. Place all dishes in an incubator (37°C with 5% CO₂) for at least 30 min before use.

    • i.
      Four dishes with 5 mL of M2 medium.
    • ii.
      One dish with 5 mL of M2 medium containing 300 μg/mL Hyaluronidase.
    • iii.
      Two dishes with 200 μL of M16 medium placed in the center and covered with 5 mL of mineral oil.
  • b.
    Euthanize the plugged females from step 1 and harvest the ovaries and oviducts. Transfer the tissues to one of the 35 mm dishes containing M2 medium.
  • c.
    Transfer one pair of ovaries and oviducts at a time to the dish containing M2 medium with Hyaluronidase. Using microdissection forceps, carefully tear open the ampulla (the swollen area of each oviduct) to release the eggs into the medium.
  • d.
    Rinse the eggs three times in M2 medium prewarmed to 37°C.
  • e.
    Transfer the eggs to M16 medium, and leave in the incubator (37°C with 5% CO₂) until ready for microinjection.
• 3.
  Microinject Cas9 mRNA and sgRNAs into mouse zygotes.

  • a.
    Prepare a total of 10 μL injection solution, containing 100 ng/μL of Cas9 mRNA and 25 ng/μL of each sgRNA in 10 mM Tris HCl, pH 7.5, 0.1 mM EDTA (prepared with RNase- and DNase-free water).

    Optional: Centrifuge the solution at >15,000 × g for 5 min at 4°C and transfer the supernatant to a fresh Nuclease-free tube. This helps remove any insoluble materials that may clog the injection needle.

  • b.
    Prepare an injection needle with a micropipette puller (Sutter Instrument P-2000, setting: Heat = 350 FIL = 4 VEL = 100 DEL = 150 PUL = 100).
  • c.
    Back load 1–2 μL of the injection solution into the needle.
  • d.
    Transfer about 50 fertilized eggs into the injection chamber, a single concave depression on a glass slide that is filled with preheated M2 medium.
  • e.
    Examine the zygotes at a high magnification. Make sure that two pronuclei are clearly visible and are morphologically normal (Figure 5). Otherwise, discard the egg.
  • f.
    Position the tip of the holding pipette next to a zygote, and apply a negative pressure to the pressure control unit.
  • g.
    Focus the microscope onto the pronuclei.
  • h.
    Set the automatic microinjector to continuous flow mode at a pressure of 40 hPa, and lower the injection needle into the M2 medium in the injection chamber.
  • i.
    Inject the RNA mix into the cytoplasm of the zygote (Figure 5).

    Note: mRNA and gRNA that are injected into the cytoplasm can efficiently enter either pronucleus.¹⁶ If injecting gRNA and Cas9 protein complexes or injecting DNA that produces Cas9 and gRNA, insert the needle into the bigger pronucleus.^17,18

  • j.
    Transfer all microinjected zygotes into the culture dish with pre-incubated, fresh M16 medium.
  • k.
    Culture the zygotes for ∼16 h in the incubator (37°C with 5% CO₂). The zygotes are expected to develop into two-cell stage the next day.
• 4.
  Transfer the microinjected zygotes into pseudo-pregnant recipient CD1 females.

  • a.
    On the same day of microinjection, set up mating pairs that each consists of two estrous CD1 females and one vasectomized CD1 male.

    Note: We typically set up 25 mating pairs for each injection.

  • b.
    Check for the presence of a vaginal plug the next morning.
  • c.
    Transfer 20–30 microinjected zygotes that have developed into two-cell embryos into the oviducts of each plugged, pseudo-pregnant CD1 female.
• 5.Continue to monitor the surrogate females until they give birth (in ∼19 days).

Note: From 10 donor females, we typically obtain ∼150 zygotes after superovulation. After microinjection and ∼16 h culturing, about half of them (i.e., 75 zygotes) develop into two-cell embryos. The embryos are then transferred into 3–4 surrogate CD1 females.

Figure 5.

Pronuclear injection of Cas9 mRNA and sgRNA

The RNA molecules are microinjected into the cytoplasm of a zygote at the pronuclear stage.

Identify F0 mice that harbor genomic mutations and clone the mutant DNA for sequencing

Timing: 1 week

The following steps produce individual plasmid clones of the genomic region of interest from the F0 pups, and to identify any mutations by Sanger sequencing.

• 6.
  Design PCR primers to amply genomic sequence spanning the gRNA target area.

  • a.
    Identify unique restriction sites on either side of the gRNA target sequence in the minigene reporter (Figure 6). For the Dcc minigene, PacI and ApaI are 5′ and 3′, respectively, to the target splice acceptor sites. troubleshooting 5.

    Note: The restriction sites are used to remove the wild-type sequence from the minigene so that it can be replaced by a mutant one. Thus non-unique sites can also be used, as long as they cut away the wild-type sequence.

  • b.
    Design forward and reverse primers that are 5′ and 3′, respectively, to the restriction sites and are at least 20 nt away from the sites.

    Note: We use recombination-based cloning strategy (i.e., NEBuilder Hifi DNA Assembly) to introduce a mutant sequence into the minigene reporter. The 20 nt sequences are the minimal overlapping arms between the insert and linearized vector, to allow efficient recombination (see Figure 6).

    Optional: If the distance between the two primers is over 1.5 kb, a third primer that anneals closer to the gRNA target sequence may be needed to facilitate Sanger sequencing.

  • c.
    Have the primers synthesized (e.g., at IDT).
• 7.
  Amplify genomic DNA sequences from all animals that are born after pronuclear injection.

  Note: As Cas9 editing efficiency is high, we typically examine 1–2 litters of F0 pups (5–10 pups per litter) for each gRNA injection.

  • a.
    Collect toe clipping samples from postnatal day 5 (P5) pups into numbered tubes.

    Note: Users may collect other tissue samples (e.g., ear punch and tail samples) based on their approved IACUC protocol. We recommend using PCR strip tubes to collect the tissue samples, so that the following steps can be performed using a thermocycler. We also recommend using multi-channel pipettes to reduce hand-on time.

  • b.
    Add 100 μL of 50 mM NaOH (diluted from 1 M NaOH with water, use within 30 days) to each tube and lyse the tissue samples by heating at 95°C for 10–15 min. Neutralize the lysate by adding 10 μL of 1 M Tris, pH 6.8. Vortex to mix thoroughly.

    Note: We do not purify the genomic DNA from this step, and use Terra polymerase to directly amplify genomic sequences (up to 2 kb) from the crude lysate. When amplifying longer genomic segments, we recommend purifying the DNA template and use high fidelity DNA polymerase, as described above for generating the minigene reporter. Store the crude lysate with any incompletely dissolved tissues at −20°C. If needed, the tissue can be lysed again with fresh 50 mM NaOH.

    Alternatives: Users may use their preferred genomic DNA extraction method and PCR reagent.

  • c.
    Assemble PCR mix using the crude lysate and the primers from step 6 (see materials and equipment).
• 8.
  Clone the amplified genomic sequences and identify mutant clones by Sanger sequencing.

  Optional: An ApaI restriction site is present within Dcc_S gRNA target (see Figure 2B), and mutations in this area are likely to eliminate the site. Thus we use ApaI digestion to pre-screen the amplicons before cloning (see materials and equipment).

  Note: WT genomic DNA is cut into two smaller fragments by ApaI, whereas mutant DNA remains as one uncut piece.

  • a.
    Run the PCR products from step 7 (after ApaI digestion for Dcc_S transgenic line) in an agarose gel (we use 2% gel for the 0.4 kb amplicon).

    Note: For Dcc_S transgenic line, proceed with the protocol using samples only from animals where half or all amplified genomic DNA is undigested by ApaI. For Dcc_L transgenic line, proceed with all samples.

  • b.
    Excise the full-length DNA band and purify the DNA with QIAquick Gel Extraction Kit.

    Note: Most mutations are small indels, so the mutant amplicons cannot be easily separated from the wild type, full-length product in gel electrophoresis. If there are additional bands that are visibly bigger or smaller than the full-length band (but are distinct from the ApaI-digested fragments), excise the additional bands and purify the DNA separately.

  • c.
    Clone the purified amplicons using TOPO TA Cloning Kit for Subcloning following the manufacturer’s protocol. After transforming the reaction into chemically competent bacteria (e.g., TOP10), incubate the LB agar plates with bacteria for ∼16 h at 37°C.

    Alternatives: For blunt-end amplicons, use blunt end TOPO cloning kit (e.g., Zero Blunt TOPO PCR Cloning Kits).

  • d.
    For the Dcc_L transgenic line, pick 6–8 colonies from each LB agar plate for direct colony sequencing (e.g., at Eurofins Genomics). For the Dcc_S line, pick 3–4 colonies for sequencing. The pCR2.1 vector in the TOPO cloning kit contains M13 and T7 primers for sequencing.

    Note: Assuming that the mutant and wild-type copies of the genomic DNA from a heterozygous animal are amplified and cloned at equal frequencies, the chance of 6–8 independent colonies all containing the wild-type sequence is 1.5%–0.4%. Thus, surveying 6–8 colonies should be sufficient to detect a heterozygous mutation. For the same reason, if the maternal and paternal alleles harbor two independent mutations, as we have previously reported,¹ picking 6–8 colonies should be sufficient to detect both mutations. For the Dcc_S transgenic line, where the wild-type allele has been digested and removed, sequencing 3–4 colonies should be sufficient to detect the mutant. If both alleles of the genomic DNA appear undigested (thus both alleles contain mutations), sequence 6–8 colonies.

    Alternatives: Start a liquid culture of single bacterial colonies, purify the plasmid DNA, and submit the plasmids for Sanger sequencing.

  • e.
    Align the sequencing results with the wild-type genomic DNA, and identify any mutant clones (we use DNASTAR software for sequence analyses). troubleshooting 6.
  • f.
    Start a liquid culture of the identified mutant colonies, culture for ∼16 h at 37°C, and purify the DNA with QIAprep Spin Miniprep Kit.

Figure 6.

Construction of a mutant minigene reporter

The wild-type sequence encompassing the targeted splicing sites is removed from the minigene plasmid by restriction digestion with PacI (5′) and ApaI (3′). An identified mutant sequence that is cloned into pCR2.1 is digested out of the vector and subcloned into the minigene plasmid using a recombination reaction.

Screen for mutations that produce the desired alternative splicing pattern(s)

Timing: 2 weeks

The following steps generate mutant minigene reporters by replacing the wild-type sequence with the identified genomic mutations from the F0 pups. The production of alternative isoforms from the mutant reporters is examined using in vitro splicing assays, in direct comparison with the wild-type reporter.

• 9.
  Subclone the identified genomic mutations into the minigene reporter.

  • a.
    Digest the wild-type minigene with the restriction enzymes identified in step 6a, i.e., PacI and ApaI for Dcc (Figure 6). Incubate at 37°C for 1 h.

    Note: Prepare the linearized plasmid in a large quantity (e.g., 5–10 μg), as it will be used for generating all mutant reporters.

  • b.
    Digest the sequenced mutant DNA clones (in pCR2.1 vector) with EcoRI.

    Note: EcoRI is present in the vector at both 5′ and 3′ ends of the inserted PCR product. If EcoRI digests the insert, additional restriction sites are available in the vector. If none of the sites are suitable, PCR amplify the insert using the same primers as in step 6a and a high-fidelity DNA polymerase.

  • c.
    Run the digestion reactions in an agarose gel. Excise the minigene vector backbone (with the wild-type target sequence deleted), and the mutant genomic DNA insert (out of the pCR2.1 vector).
  • d.
    Purify the vector and insert fragments with QIAquick Gel Extraction Kit.
  • e.
    Set up a recombination reaction using NEBuilder HiFi DNA Assembly Master Mix (see materials and equipment). Include a negative control where only the linearized vector is present in the recombination reaction.
  • f.
    Transform 2 μL of the recombination reaction into 50 μL of chemically competent bacteria. After heat shock and recovery, spread the bacteria on an agar plate with 100 μg/mL ampicillin or 50 μg/mL kanamycin.

    Note: If the negative control yields many colonies, repeat the restriction digestion (step a) by extending the digestion time or increasing the enzyme amount.

  • g.
    Select a few colonies to start a 5 mL liquid culture and purify the DNA using QIAprep Spin Miniprep Kit.
  • h.
    Determine if the clones contain the identified mutations by Sanger sequencing.

    Note: These clones are the mutant minigene reporters.

• 10.
  Compare the splicing patterns between the wild-type and mutant minigene reporters in cultured cells (following the same procedure outlined in steps 10–14 of before you begin for the wild-type reporter).

  • a.
    Subculture COS-1 cells and plate the cells in a 24 well plate.

    Note: The number of wells is determined by the number of identified mutations.

  • b.
    The next day, transfect the cells with the wild-type or mutant minigene reporter.

    Note: If relevant, include conditions where splicing regulators, e.g., Nova1 and Nova2, are co-expressed for comparison.

  • c.
    24–48 h post-transfection, lyse the cells with direct lysis buffer for RNA extraction.
  • d.
    Perform reverse transcription using a reverse primer (e.g., T7 for the Dcc minigene) and first-strand cDNA synthesis reagents.
  • e.
    Perform semi-quantitative PCR using the same primers and cycle number as determined for the wild-type minigene reporter.
  • f.
    Analyze the amplicons with agarose gel electrophoresis and identify the mutations that produce the desired splicing patterns (Figure 4B). troubleshooting 7.
• 11.Identify the F0 mice that harbor the desired mutations and set up outcrossing breeding to introduce the mutations into the desired genetic background.

Note: The genetic background of the F0 mice is the strain used for pronuclear injection, i.e., FVB/NJ in this protocol, not that of the surrogate females.

CRITICAL: Perform multiple generations of outcrossing before detailed phenotypic analyses to minimize artifacts from the genetic background or from off-target mutations.

• 12.Design primers for PCR-based genotyping to distinguish the mutant and wild-type alleles.

Note: For small deletion mutations, we recommend designing a primer that recognizes the cojoining ends spanning the deleted sequence; see Figure 7).

CRITICAL: We have found that small indels can be occasionally repaired during outcrossing or backcrossing. For example, we have found that an original 14 bp deletion was reduced to 4 bp within a year of backcrossing with the C57BL/6J wild-type strain. Thus, it is critical to use mutation-specific oligos for genotyping. We also recommend that the mutations be regularly monitored by Sanger sequencing.

Figure 7.

Design of genotyping primers

A 9 bp sequence (in light gray) is deleted near Dcc_L splice site (in red letters) in the mutant shown, which leads to the sole expression of Dcc_L. Primer sequences are underlined. The 9 bp deletion-specific reverse primer recognizes the conjoining ends of the sequence after deletion. When all three primers are included for PCR-based genotyping, annealing of the mutant primer with the deletion allele blocks the wild-type reverse primer from generating the wild-type products (i.e., the wild-type forward and reverse primers produce only a product from the wild-type allele).

Expected outcomes

Some lethality is expected in F0 mouse pups that are born after pronuclear injection and embryo transfer. Besides having the procedures done by experienced researchers, pairing a recipient female with a second pseudo-pregnant female may help improve the care of neonates. In our experience, half or more of the F0 pups survive until adulthood.

Among the survived F0 mice, the frequency of genome editing by Cas9 is expected to be high. In our experience, over 80% of F0 mice harbor at least one allele of genomic mutations. Occasionally, both alleles are mutated in a single animal. This confirms that the Cas9 mRNA and gRNA that are injected into the cytoplasm of zygotes can efficiently enter either or both pronuclei and induce independent DNA breaks and repairs.¹⁶ As previously reported, the majority of mutations are expected to be in close proximity to the NGG PAM sequence.⁷

We expect that disrupting the surrounding sequences of a consensus splice site will alter the usage of the site. The resulting alternative splicing pattern will need to be empirically determined. We expect that the splicing patterns produced from the wild-type and mutant minigene reporters are highly consistent between in vitro and in vivo observations. In other words, the cell culture based splicing assays can accurately predict the resulting splicing pattern from a genomic mutation. Thus, the in vitro assay is a reliable and valuable way to quickly screen for the desired mutations.

Limitations

In the case of the two splice acceptor sites at the intron16/exon 17 boundary of the Dcc locus, we are able to alter the usage of the site by introducing small indels in close proximity.¹ Specifically, removing a stretch of apparently GC-rich sequence leads to the exclusive usage of each of the splice site.¹ However, we do not know if this finding can be extrapolated to other splice sites. We also do not know if targeting a distinct region can similarly alter the splicing pattern. As we have not identified all elements of regulatory sequences for the Dcc splicing sites, our approach relies on screening random indel mutations, instead of on introducing defined replacement sequences.

Troubleshooting

Problem 1

No gRNA candidates are identified in close proximity to the target splice site (in step 1 of before you begin).

Potential solution

Wild-type SpCas9 requires the presence of adjacent NGG PAM sequence to bind to the target DNA. If NGG PAM is not available in the vicinity of the desired cut location, one may consider using engineered variants of SpCas9 (e.g., SpCas9-NG^19,20) or other Cas nucleases (e.g., Cas12a²¹) that recognize different PAM sequences.

Problem 2

No PCR product is amplified from phenol/chloroform-extracted genomic DNA (in step 8 of before you begin).

Potential solution

One may consider using higher quality genomic DNA that contains less PCR inhibitors than phenol/chloroform-extracted DNA, such as ready-to-use genomic DNA from a commercial source or DNA purified using a column-based kit (e.g., DNeasy Blood and Tissue Kit from QIAGEN). Alternatively, one may consider using DNA polymerases that are more tolerant of PCR inhibitors, such as Terra polymerase (as Terra is not a high-fidelity polymerase, it is suitable for shorter, up to ∼2 kb amplicons).

If the problem persists, one may consider amplifying smaller, overlapping fragments of genomic DNA and piecing them together using recombination reactions (e.g., NEBuilder Hifi DNA Assembly and NEBridge Golden Gate Assembly). As many as 6 fragments, with 20–30 bp overlaps between fragments can be seamlessly assembled with NEBuilder Hifi DNA Assembly. Up to 30 fragments with 3–4 bp overlaps can be assembled with NEBridge Golden Gate Assembly.

Problem 3

The genomic region spanning the alternatively spliced sequence (i.e., the minigene) is too big to clone into a plasmid (in step 8 of before you begin).

Potential solution

In our previous study of the Robo1 and Robo2 genes, the alternatively spliced exon is flanked by introns that are ∼25 kb in total. Instead of cloning the complete sequence of the flanking exons and introns, we included only ∼600 bp from the 5′ and 3′ introns each in the minigene reporter. We show that the resulting reporter is able to recapitulate the splicing pattern in cultured cells as in vivo (see Johnson et al.² for details). We selected the 600 bp intron sequences based on the fact that they contain previously identified binding sites for NOVA splicing factors,² which are required to regulate the alternative splicing event in vivo.

Thus, if the genomic DNA fragment is too big to clone into a plasmid, we recommend making serial truncations of flanking introns to determine the minimally required regions to include in the minigene reporter. When introducing such truncations, recombination reactions (e.g., using NEBuilder Hifi DNA Assembly) allow seamless connections of DNA fragments without the need for restriction digestion or ligation.

Problem 4

No alternative isoforms are produced from the minigene reporter in COS-1 cells. Or the splicing pattern obtained from the minigene is different from that in vivo (in step 14 of before you begin).

Potential solution

COS-1 cells may not express the same splicing regulators as in the tissue of interest, thus the minigene may not fully recapitulate the in vivo splicing pattern. Users can compare different transfectable cell lines or primary cells to select a cell type that regulates the splicing event similarly as in vivo. It may also help to co-transfect the relevant splicing factors, if known. However, it is not critical that the ratio between the alternative isoforms is the same in cultured cells as in vivo, as long as the in vitro assay can detect changes induced by the identified mutations.

Problem 5

There are no restriction sites 5′ or 3′ to the target area in the minigene reporter (related to step 6).

Potential solution

PCR-based mutagenesis, using synthesized oligos containing the identified mutations, can be performed alternatively to replace the wild-type sequence in the minigene. Use high-fidelity DNA polymerase to avoid introducing unwanted mutations.

Problem 6

No mutants are identified after pronuclear injection (related to step 8).

Potential solution

First, validate the Cas9 expression construct with whole plasmid sequencing and make sure there are no missense mutations or premature stop codons. Second, identify any additional ATG sequences that may be present 5′ to the start codon of Cas9. Make sure all ATG sequences are the in the same reading frame. Third, users can also take advantage of commercially available Cas9 mRNA or protein.

Alternatively, users may consider selecting a different gRNA with a higher on-target score/ranking (even if the off-target probabilities are also increased). Upon identifying the desired mutations, users can minimize potential off-target mutations with extensive outcrossing/backcrossing breeding.

Problem 7

Unexpected or additional alternative splicing events are observed from a mutant minigene reporter (related to step 10).

Potential solution

When a canonical splice site is disrupted, cryptic splice sites may be used instead. The indel mutations may also introduce new cryptic splice sites. We have observed the emergence of two cryptic sites when we introduced single nucleotide changes around Dcc_S splice acceptor site.³ Thus, it is critical to use the in vitro splicing assay to determine the actual outcomes of the genomic mutations.

Resource availability

Lead contact

Requests for further information, resources, and reagents should be directed to the lead contact, Dr. Zhe Chen (chen6867@umn.edu).

Technical contact

Technical questions on performing the protocol should be directed to the technical contact, Dr. Zhe Chen (chen6867@umn.edu).

Materials availability

This study did not generate new unique reagents.

Data and code availability

This study did not generate datasets or original codes.

Acknowledgments

This work was supported by the following grants: Masonic Institute for the Developing Brain grant (Z.C.), National Institutes of Health (NIH) R21DA056728 (Z.C.), and National Institutes of Health (NIH) R01EY024261 (H.J.) and R01EY033316 (H.J.). We thank the MCDB transgenic facility at the University of Colorado Boulder for generating CRISPR-Cas9-mediated mutant mice. The graphical abstract was created with BioRender.com.

Author contributions

Conceptualization, Z.C.; development of the protocol, Y.T., K.A., and Z.C.; writing – original draft, Z.C.; writing – editing, Y.T., K.A., H.J., and Z.C.

Declaration of interests

The authors declare no competing interests.