CRISPR Knockout Protocol: From gRNA Design to Validated Cell Line

Overview of the CRISPR Knockout Workflow

A CRISPR knockout (KO) experiment is best understood as a sequence of discrete, checkpointed stages rather than a single experiment. Each stage has its own failure modes, and decisions made early — particularly in gRNA design and delivery method selection — have compounding effects on downstream validation. The full workflow from design to a banked, validated clone typically takes 8–16 weeks in a mammalian cell line, depending on cell doubling time and the depth of validation required.

The stages, in order, are:

1. Guide RNA (gRNA) design and off-target analysis
2. Delivery method selection and vector/reagent preparation
3. Transfection and early enrichment or selection
4. Single-cell cloning
5. Genotypic validation (Sanger sequencing / ICE analysis)
6. Phenotypic validation (RT-qPCR and Western blot)
7. Banking and functional confirmation

Each is addressed in detail below, with approximate timing per step.

Step 1: Guide RNA Design and Off-Target Analysis

Timing: 1–3 days (design); 3–5 days if synthesizing sgRNA oligonucleotides in-house.

The single guide RNA (sgRNA) directs Cas9 to a 20-nucleotide target sequence immediately upstream of a 5′-NGG protospacer adjacent motif (PAM). For a clean knockout, target the earliest constitutively expressed exon — typically exon 1 or the first coding exon — to maximize the probability that NHEJ-induced frameshift mutations produce a null allele via nonsense-mediated mRNA decay (NMD) or truncation of a critical domain.

Design Principles

• Exon selection: Target an exon present in all known transcript isoforms. Check Ensembl or UCSC Genome Browser for alternative transcription start sites before committing to a locus.
• On-target efficiency scoring: Use at least two orthogonal scoring algorithms — Doench Rule Set 2 (azimuth) and CRISPOR are well-benchmarked against empirical editing data. Prioritize guides with on-target scores ≥ 0.5 (Rule Set 2 scale).
• GC content: Aim for 40–70% GC in the 20-nt spacer. Poly-T stretches of four or more consecutive thymines should be avoided when using U6-driven expression, as they function as Pol III termination signals.
• Design two to four independent gRNAs per gene to hedge against activity variation. Rank by on-target score, then filter by off-target profile.

Off-Target Analysis

Off-target editing is a real concern, particularly in safety-critical or publication-bound projects. Run each candidate guide through Cas-OFFinder or CRISPOR's CFD (Cutting Frequency Determination) off-target scorer. Flag any predicted off-target site with ≤ 2 mismatches in the seed region (positions 1–12 from the PAM). For high-stringency projects, whole-genome sequencing of the final clone, or targeted amplicon sequencing at the top five predicted off-target loci, is appropriate.

Selecting a high-fidelity Cas9 variant — eSpCas9(1.1), HiFi Cas9, or HypaCas9 — measurably reduces off-target activity with only modest reduction in on-target efficiency in most published comparisons.

Step 2: Delivery Method Selection

Timing: Variable — RNP assembly is same-day; lentiviral production adds 7–10 days.

Three delivery formats are in common use, each with distinct trade-offs in editing efficiency, off-target risk, cargo capacity, and cost.

RNP Delivery — The Preferred Format for Most Projects

Electroporation of preassembled Cas9:sgRNA RNP complexes is the current standard for most mammalian knockout projects. The Cas9 protein turns over within 24–48 hours post-delivery, substantially limiting the window for off-target cleavage. RNP complexes are assembled by incubating recombinant Cas9 protein with in vitro-transcribed or chemically synthesized sgRNA (typically at a 1:1.2 to 1:2 molar ratio) for 10–15 minutes at room temperature before electroporation.

Electroporation conditions (voltage, pulse duration, cuvette format) must be optimized per cell line. Nucleofector-based systems (Lonza) and the Neon Transfection System (Thermo Fisher) are commonly used. Lipofection of RNPs is viable for some adherent lines but is less efficient than electroporation in most settings.

Plasmid Delivery

All-in-one plasmids (e.g., pX459 from Addgene) co-express Cas9 and the sgRNA from a single construct and include a puromycin resistance cassette for transient enrichment of transfected cells. The main disadvantage is prolonged Cas9 expression from episomal DNA, which correlates with increased off-target editing frequency compared to RNP delivery in multiple published datasets.

Lentiviral Delivery

Lentiviral transduction is used when the target cell type is refractory to lipofection or electroporation — primary T cells, induced pluripotent stem cells (iPSCs), and post-mitotic neurons are the most common examples. Stable Cas9 integration allows inducible or sequential editing but requires careful characterization of integration sites and sustained monitoring for off-target activity.

Step 3: Transfection and Early Enrichment

Timing: Day 0 (transfection) through Day 3–5 (selection or sorting).

Immediately after transfection or transduction, allow cells to recover for 24–48 hours in complete medium without selection. For plasmid-based delivery using a resistance cassette, apply puromycin (or the appropriate antibiotic) for 48–72 hours at a concentration titrated to kill untransfected cells within 3–5 days. This transient selection enriches the edited pool before clonal expansion — it does not guarantee a KO, as many puromycin-resistant cells will carry wild-type alleles.

For RNP delivery, where no selection marker is present, fluorescence-activated cell sorting (FACS) can enrich cells co-electroporated with a GFP-encoding mRNA or plasmid. Alternatively, proceed directly to single-cell cloning from the bulk-edited pool after confirming bulk editing efficiency by a T7 Endonuclease I (T7EI) assay or Sanger/ICE analysis of a pooled sample.

Bulk pool check: Before investing in clonal expansion, confirm that the bulk edited population shows ≥ 30% indel frequency by ICE or TIDE analysis. If editing efficiency is below this threshold, re-optimize delivery conditions before cloning.

Step 4: Single-Cell Cloning

Timing: 2–4 weeks for colony outgrowth, depending on cell doubling time.

A validated CRISPR knockout cell line protocol requires isolation of clonal populations to obtain cells with defined, homozygous (or compound heterozygous) frameshift mutations on all alleles. Biallelic disruption is required for functional knockout of most genes; monoallelic editing leaves one wild-type copy sufficient for protein production.

Cloning Methods

• Limiting dilution: Seed 0.3–0.5 cells/well in 96-well plates in conditioned medium. Mark wells with a single colony at 7–10 days and expand. Simple, no specialized equipment required.
• FACS single-cell sorting: More reliable for verifying single-cell origin, particularly for rapidly growing lines prone to satellite colony formation during limiting dilution. Sorting into 96-well plates pre-loaded with conditioned medium or feeder cells improves outgrowth.
• Semi-solid medium (methylcellulose or agarose-based): Used for suspension cells (e.g., Jurkat, K562) where limiting dilution and FACS are less efficient.

Expand colonies in parallel tracks — keep a replica plate at an earlier passage as a backup. Do not passage too aggressively before genotyping; slow-growing clones are easily lost.

Step 5: Genotypic Validation — Sanger Sequencing and ICE Analysis

Timing: 3–7 days from DNA extraction to data.

Genotypic validation confirms that the genomic sequence at the target locus is disrupted on all alleles. This is a necessary but not sufficient condition for a functional knockout — it does not confirm loss of protein expression.

Sanger Sequencing at the Target Locus

1. Extract genomic DNA from each clone (column-based kits are adequate for 96-well scale).
2. PCR-amplify a 400–600 bp amplicon centered on the cut site. Design primers 150–200 bp outside the expected indel region to avoid primer binding interference with large deletions.
3. Submit PCR product for Sanger sequencing with a primer that reads cleanly through the cut site.
4. Analyze the resulting trace using the ICE (Inference of CRISPR Edits) tool (Synthego) or TIDE (Tracking of Indels by Decomposition). Both tools deconvolve mixed Sanger traces from cells with two or more distinct alleles into individual allele calls.

Confirm that all alleles carry frameshift-inducing indels (insertions or deletions not divisible by 3). In-frame indels of small size (e.g., ±3 bp) should be flagged — they delete or insert single amino acids but may not ablate protein expression.

Subcloning the PCR amplicon into a plasmid and sequencing multiple colonies is more laborious but provides unambiguous allele-level resolution in polyploid cell lines or lines with complex karyotypes (e.g., many cancer cell lines).

Step 6: Phenotypic Validation — RT-qPCR and Western Blot

Timing: 5–10 days, depending on antibody availability and protocol optimization.

This is where many CRISPR knockout projects fail — not because the editing did not occur, but because validation at the protein level is either skipped or conducted with an antibody that is not specific enough to detect residual expression. Genotypic confirmation of a frameshift is necessary evidence, but it is not sufficient to publish or to use the line in a functional assay. Protein-level confirmation is required.

RT-qPCR

RT-qPCR detects changes in mRNA abundance. In a clean knockout, NMD typically degrades the frameshifted transcript, reducing mRNA levels by 50–90% relative to wild-type controls. Design qPCR primers spanning an exon-exon junction downstream of the cut site to avoid amplifying genomic DNA and to detect the mature transcript specifically. Normalize to at least two stably expressed reference genes (e.g., GAPDH, HPRT1, or ACTB — validate reference gene stability in your specific cell line).

Note that partial mRNA reduction is not definitive evidence of complete KO. Some frameshifted transcripts escape NMD, and alternative translation initiation can produce truncated proteins. Always proceed to Western blot.

Western Blot

Western blot is the standard final confirmation step in any CRISPR knockout validation strategy. It directly measures protein abundance and — when interpreted alongside genotyping data — provides the most direct evidence that the gene product is absent.

Critical considerations for Western blot validation:

• Antibody specificity: The antibody must be validated against the target protein in the relevant cell line background. An antibody showing cross-reactivity to a related family member will produce a false-positive band even in a genuine KO. Use isotype controls and, ideally, validate antibody specificity using a KO lysate as a negative control — which is circular logic unless an orthogonal antibody is used. This is why matched, pre-validated antibodies matter.
• Expected molecular weight: Confirm the detected band in wild-type cells corresponds to the annotated molecular weight of the target. Run known positive control lysates alongside experimental samples.
• Loading control: Probe for a housekeeping protein (e.g., β-actin, GAPDH, or vinculin) on the same or stripped membrane. Identical loading controls confirm that absence of target signal is not an artifact of underloading.
• Long exposure: Run both a standard and an extended exposure to confirm absence of low-abundance residual protein.

At Triple Point Biologics, every custom knockout cell line is validated by Western blot using a matched, characterized antibody — not a randomly selected commercial reagent — because reagent quality at this step determines whether the validation data are interpretable. TPB has produced characterized proteinase and inhibitor antibodies since 1994, and this depth of antibody validation experience informs how KO validation is approached.

The CRISPR Knockout Validation Gap

In practice, the single most common reason CRISPR knockout projects fail to produce publishable data is incomplete validation. Studies that confirm only the genotypic level (Sanger + ICE analysis) without protein-level Western blot data, or that use antibodies of insufficient specificity, leave open the possibility of residual protein expression from alternative start sites, read-through, or upstream regulatory rescue. Reviewers at major journals increasingly require both genotypic and protein-level evidence for KO claims.

A secondary failure mode is selection of clones with heterozygous edits — one disrupted allele and one wild-type — due to insufficient clonal expansion or early passaging. Rigorous ICE analysis of multiple independent clones, combined with Western blot of all validated clones, is the minimum standard for a publication-quality KO line.

CRISPR Knockout Validation Checklist

• ☐ Bulk pool ICE analysis confirms ≥ 30% indel frequency before single-cell cloning
• ☐ Sanger sequencing of target locus in ≥ 3 independent clones
• ☐ ICE or TIDE deconvolution confirms frameshift on all alleles
• ☐ In-frame indels flagged and excluded, or further characterized
• ☐ RT-qPCR shows mRNA reduction in KO clones relative to wild-type control
• ☐ Western blot with validated, target-specific antibody shows absent band in KO clones
• ☐ Loading control confirms equal protein input across lanes
• ☐ At least two independent KO clones show concordant phenotype (reduces probability of off-target artifacts driving the phenotype)
• ☐ Off-target loci sequenced if project is safety-relevant or phenotype is unexpected
• ☐ Cells banked at low passage number after full validation

Troubleshooting Common Problems

Low or No Editing Efficiency in the Bulk Pool

• Confirm cell viability post-electroporation; dead cells skew ICE traces. Target ≥ 70% viability at 24 hours post-transfection.
• Re-optimize electroporation parameters (voltage, pulse width, cell density).
• Switch to a different gRNA if efficiency remains below 20% after two rounds of optimization — not all genomically predicted guides are active in cells.
• Confirm that the target locus is not in a region of low chromatin accessibility. ATAC-seq data or DNase I hypersensitivity data for the target cell line (available in ENCODE) can guide locus selection.

All Clones Are Heterozygous

• This commonly reflects haploinsufficiency — loss of both alleles reduces cell fitness and heterozygous clones outcompete nulls during expansion. If the gene is essential, consider an inducible knockout system (Tet-on Cas9 or auxin-inducible degron).
• Alternatively, increase the number of clones screened. Screening 96+ clones per experiment, rather than 24, substantially increases the probability of recovering biallelic KO clones for most genes.

Western Blot Shows Band in KO Clones

• First, verify antibody specificity using an orthogonal approach (e.g., a second antibody raised against a different epitope).
• Check whether an alternative translation initiation codon downstream of the cut site could produce a truncated but immunoreactive protein.
• Revisit ICE analysis — verify that both alleles in the clone are genuinely frameshifted, not in-frame.
• Confirm that the cell line is not pseudodiploid with more than two copies of the target locus (relevant for aneuploid cancer cell lines).

Inconsistent RT-qPCR Results Across KO Clones

• Validate reference gene stability across your KO and wild-type clones using the geNorm or NormFinder algorithm. Loss of a target gene can alter expression of commonly used reference genes in some contexts.
• Ensure qPCR primers amplify downstream of the indel to avoid amplifying a truncated transcript that retains the primer binding sites but is functionally null.

Timing Summary by Stage

When to Use a Custom KO Service vs. In-House Production

Generating a validated CRISPR knockout cell line in-house is feasible for labs with established electroporation workflows, FACS access, and characterized antibodies for the target. The resource investment is non-trivial: staff time for 10–16 weeks, reagent costs, and — critically — access to a validated antibody for Western blot confirmation.

A contracted custom KO service is worth considering when:

• The target cell line is difficult to transfect (primary cells, suspension lines, iPSCs)
• The lab lacks a validated antibody for the target protein and does not have the resources or timeline to develop one
• Multiple gene knockouts are needed in parallel
• The project requires a documented, audit-ready validation package
• In-house FACS for single-cell sorting is not readily available

For labs considering a combined antibody + KO validation approach, see the TPB guide on protease activity assays for context on how protein-level functional validation integrates with cell line characterization. For catalog reagents relevant to target confirmation, see the TPB recombinant protein collection.

Frequently asked questions