What is Dr. Zubair Khalid's research focus?

Dr. Zubair Khalid specializes in molecular virology, mRNA vaccine development, and computational biology, with a focus on avian pathogens like IBDV and Avian Reovirus.

Where is Dr. Zubair Khalid currently working?

Dr. Zubair Khalid is a Postdoctoral Research Associate at the University of Maryland (UMD), specifically within the Department of Animal and Avian Sciences.

How to Use CRISPR: A Step-by-Step Guide for Genome Editing

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-11

Introduction to CRISPR: What It Is and How It Works

CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) is an adaptive immune system naturally found in bacteria and archaea. When a bacteriophage infects a bacterium, the bacterial cell can capture a short fragment of the viral DNA and integrate it into its own genome at a CRISPR locus. This stored sequence serves as a molecular memory. Upon subsequent infection by the same phage, the bacterium transcribes the CRISPR locus into RNA, processes it into short CRISPR RNAs (crRNAs), and uses these to guide a nuclease to the matching viral DNA, which is then cleaved and destroyed.

The scientific breakthrough came when researchers realized this system could be reprogrammed to cut any DNA sequence of choice. The most widely used version, CRISPR-Cas9, is derived from Streptococcus pyogenes. It consists of two core components: a Cas9 endonuclease and a single guide RNA (sgRNA). The sgRNA is a synthetic fusion of the crRNA (which carries the target-specific sequence) and a tracrRNA (trans-activating crRNA) that is required for Cas9 to become catalytically active. By simply changing the 20-nucleotide spacer sequence at the 5′ end of the sgRNA, you can direct Cas9 to essentially any genomic locus that is followed by a protospacer adjacent motif (PAM).

Components of the CRISPR-Cas9 System

The three essential elements for a basic CRISPR experiment are:

Cas9 nuclease: A large (~160 kDa) endonuclease with two nuclease domains, HNH and RuvC. The HNH domain cleaves the DNA strand complementary to the sgRNA, while the RuvC domain cleaves the non-complementary strand. Both domains must be active for a double-strand break (DSB) to occur. The S. pyogenes Cas9 (SpCas9) is the most common variant and recognizes the NGG PAM sequence, where N is any nucleotide.

Single guide RNA (sgRNA): A chimeric RNA molecule of ~100 nucleotides. The first 20 nucleotides at the 5′ end are the spacer, which is complementary to the target DNA sequence. The remaining ~80 nucleotides form a scaffold that binds Cas9 and is required for its activation. The spacer sequence must be exactly complementary to the target DNA for efficient cleavage; mismatches, especially in the "seed region" (the 8–12 nucleotides adjacent to the PAM), severely reduce activity.

PAM sequence: A short, conserved DNA motif immediately downstream (3′) of the target site. For SpCas9, this is 5′-NGG-3′. The PAM is essential for two reasons: it distinguishes self from non-self DNA in the bacterial immune system, and it is required for Cas9 to initially interrogate and unwind the DNA. Cas9 first binds to the PAM, then checks whether the adjacent 20-nucleotide sequence matches the sgRNA. If there is no PAM, Cas9 will not bind or cleave. This means that not every 20-nucleotide sequence in your gene of interest is targetable; you must find a sequence that is immediately followed by an NGG motif.

Mechanism of DNA Cleavage and Repair

Once the sgRNA and Cas9 form a ribonucleoprotein (RNP) complex, the complex scans the genome for PAM sequences. When Cas9 encounters an NGG PAM, it unwinds the DNA duplex and attempts to pair the sgRNA spacer with the complementary strand. If the match is sufficient, Cas9 undergoes a conformational change that activates both nuclease domains, generating a blunt double-strand break (DSB) approximately 3 nucleotides upstream of the PAM.

The cell then repairs the DSB through one of two endogenous pathways:

Non-homologous end joining (NHEJ): This is the dominant repair pathway in most cell types, especially in non-dividing or slowly dividing cells. NHEJ directly ligates the broken ends together, often introducing small insertions or deletions (indels) at the break site. These indels frequently cause frameshift mutations that introduce premature stop codons, effectively knocking out the gene. This is the basis of CRISPR Knockout experiments.

Homology-directed repair (HDR): This pathway uses a homologous DNA template to repair the break with high fidelity. HDR is active mainly in the S/G2 phases of the cell cycle. If you supply an exogenous donor template—such as a single-stranded oligodeoxynucleotide (ssODN) or a plasmid with homology arms—you can introduce precise insertions, point mutations, or epitope tags. This is the basis of knock-in experiments.

The choice between NHEJ and HDR is largely determined by the cell type, the phase of the cell cycle, and the availability of a donor template. For most gene knockout applications, NHEJ is sufficient and efficient. For precise edits, HDR must be optimized, often by using cell synchronization or chemical inhibitors of NHEJ.

Designing Guide RNAs for Your Target Gene

The success of a CRISPR experiment hinges on guide RNA design. A poorly designed sgRNA can result in no editing, high off-target activity, or both. The design process involves selecting a target sequence, verifying PAM availability, and using computational tools to score specificity.

Selecting Target Sites and PAM Sequences

For a gene knockout, you want to introduce a frameshift mutation early in the coding sequence (CDS), ideally in the first or second exon. This ensures that any truncated protein is non-functional. The target sequence should be 20 nucleotides long, immediately followed by an NGG PAM on the non-target strand. The 20-nucleotide spacer is part of the sgRNA; the PAM is not included in the sgRNA but is required for Cas9 to bind.

Consider the human HPRT1 gene, commonly used as a control for CRISPR experiments. A typical target might be 5′-GATGATGAACCAGGTATGAT-3′ followed by a PAM of 5′-TGG-3′. The sgRNA spacer would be the 20-nucleotide sequence, and the PAM is the adjacent 5′-NGG-3′ on the opposite strand.

Key design rules:

Avoid targeting the 3′ end of the gene: A truncation at the C-terminus may retain partial function.
Prefer exons that are common to all transcript isoforms: Use a database like Ensembl or NCBI to check transcript variants.
Avoid poly-T stretches: A run of four or more thymines in the sgRNA coding sequence acts as a termination signal for RNA polymerase III, which is used to express sgRNA from a U6 promoter. This will result in truncated, non-functional sgRNA.
Check for single nucleotide polymorphisms (SNPs): If your target site contains a common SNP, the sgRNA may not bind in some individuals or cell lines.

For knock-in experiments, the target site should be as close as possible to the intended insertion site. The HDR efficiency drops sharply with increasing distance between the DSB and the desired edit; aim for less than 10–20 nucleotides.

Using Online Design Tools

Several free, web-based tools automate guide design and off-target prediction. The most widely used are:

Benchling: An integrated platform that allows you to upload your gene sequence, design guides, and visualize the cutting site. It provides an on-target score and lists potential off-target sites with a specificity score.
CRISPOR: A dedicated guide design tool that integrates multiple scoring algorithms (e.g., MIT specificity score, CFD score) and provides primer recommendations for validation.
CRISPRscan: Particularly useful for designing guides for zebrafish and other model organisms, but also works for mammalian systems.

These tools work by scanning your input sequence for all possible 20-NGG sites, then aligning each candidate against the reference genome to identify off-target matches. They output a ranked list of guides with scores for efficiency and specificity. A good guide typically has an on-target score above 50 (on a 0–100 scale) and a high specificity score, with few or no off-target sites that have fewer than 3 mismatches.

A common mistake is to design a guide manually without checking off-targets. Even a single mismatch in the seed region can allow Cas9 to bind and cleave at an unintended locus. Always use a computational tool and choose the guide with the best combined score. For further reading on minimizing unintended edits, see CRISPR Off Target Effects.

Choosing the Right CRISPR System and Delivery Method

Not all CRISPR systems are identical. The choice of nuclease and delivery method depends on your cell type, the type of edit you want to make, and your tolerance for off-target effects.

Cas9 vs. Cas12a vs. Other Nucleases

The most common nuclease is SpCas9, but several alternatives offer distinct advantages:

Feature	SpCas9	Cas12a (Cpf1)	Cas9 nickase (D10A)
PAM requirement	5′-NGG-3′	5′-TTTV-3′ (V = A/C/G)	5′-NGG-3′
Cleavage pattern	Blunt ends	Staggered cuts with 4–5 nt overhangs	Single-strand nick
Guide RNA	Single sgRNA (~100 nt)	Single crRNA (~42 nt, no tracrRNA)	Single sgRNA
Target site length	20 nt	19–23 nt	20 nt
Off-target activity	Moderate	Lower	Very low (requires paired guides)
Best use case	General knockout/knock-in	AT-rich regions, multiplexing	Precision editing with minimal off-targets

Cas12a (previously known as Cpf1) recognizes a T-rich PAM (5′-TTTV-3′), which is useful for editing AT-rich genomes or regions where NGG sites are rare. It also generates staggered cuts, which can improve HDR efficiency in some contexts. Additionally, Cas12a processes its own crRNA array, making it easier to multiplex—express multiple guides from a single transcript.

The Cas9 nickase (D10A) has one of its nuclease domains inactivated, so it cuts only one strand. To create a DSB, you must use a pair of nickases targeting opposite strands within a short window (typically 20–40 bp apart). This strategy dramatically reduces off-target effects because simultaneous off-target nicks at the same locus are extremely rare. However, it requires designing two guides and is more complex to set up.

For most undergraduate and routine laboratory applications, wild-type SpCas9 is sufficient. If you are working with a difficult-to-transfect cell type or need minimal off-target activity, consider the nickase or Cas12a.

Delivery Methods: Plasmid, mRNA, RNP

There are three main ways to deliver the CRISPR components into cells:

Plasmid DNA: The Cas9 gene and sgRNA are cloned into a plasmid, often with a selectable marker (e.g., puromycin resistance) or a fluorescent reporter (e.g., GFP). The plasmid is introduced into cells via transfection or electroporation. Pros: inexpensive, easy to produce, allows selection of edited cells. Cons: expression is sustained for days, increasing off-target effects; the plasmid must enter the nucleus for transcription; some cell types are refractory to plasmid transfection.

mRNA + sgRNA: Cas9 is delivered as in vitro-transcribed mRNA, and the sgRNA is delivered separately (either as in vitro-transcribed RNA or as a chemically synthesized oligo). Pros: transient expression reduces off-target effects; mRNA does not need to enter the nucleus for transcription. Cons: more expensive than plasmids; mRNA is labile and requires careful handling (RNase-free conditions, storage at −80°C).

Ribonucleoprotein (RNP) complex: Purified Cas9 protein is complexed with the sgRNA in vitro, and the pre-formed RNP is delivered into cells. Pros: the fastest onset of activity (minutes, not hours); the complex is degraded within 24–48 hours, minimizing off-target effects; no risk of genomic integration of plasmid DNA; works well in hard-to-transfect cells like primary neurons and stem cells. Cons: the most expensive option; requires optimization of electroporation conditions.

For standard cell lines like HEK293T or HeLa, plasmid transfection with a lipid reagent (e.g., Lipofectamine 3000) is simple and effective. For primary cells, induced pluripotent stem cells (iPSCs), or in vivo work, RNP delivery via electroporation (e.g., Neon or Lonza Nucleofector) is strongly preferred.

Performing the CRISPR Experiment: Step-by-Step Protocol

The following protocol is a general guide for a CRISPR knockout in a mammalian cell line using plasmid delivery. Adjust volumes and conditions based on your specific cell type and reagents.

Cell Preparation and Transfection

Culture cells to 70–80% confluency in a 6-well plate. Use complete medium appropriate for your cell line (e.g., DMEM + 10% FBS for HEK293T). Cells should be in logarithmic growth phase; healthy, actively dividing cells take up DNA more efficiently.

Prepare the transfection mix. For one well of a 6-well plate, dilute 2.5 µg of the CRISPR plasmid (encoding both Cas9 and sgRNA) in 125 µL of Opti-MEM reduced serum medium. In a separate tube, dilute 5 µL of Lipofectamine 3000 in 125 µL of Opti-MEM. Mix the two solutions and incubate at room temperature for 10–15 minutes.

Add the mixture dropwise to the cells. Gently rock the plate to distribute the transfection complex evenly. Return the cells to the incubator at 37°C with 5% CO₂.

Change the medium after 6–12 hours. This removes any residual transfection reagent that may be toxic. If your plasmid carries a fluorescent marker, you can check transfection efficiency at 24 hours post-transfection by fluorescence microscopy. Expect 50–90% efficiency in HEK293T cells; lower in more difficult lines.

Harvest cells 48–72 hours post-transfection for DNA extraction and validation. If your plasmid has a selectable marker, you may choose to enrich for edited cells by adding the selection drug (e.g., 2 µg/mL puromycin) at 24 hours post-transfection and harvesting after 3–5 days of selection.

Optimizing Editing Efficiency

If your editing efficiency is low, consider the following adjustments:

Increase the amount of plasmid DNA (up to 4 µg per well of a 6-well plate), but be aware of toxicity.
Use a higher-efficiency transfection reagent or switch to electroporation. For HEK293T, electroporation with the Neon system (one pulse at 1,100 V for 30 ms) routinely gives >90% transfection efficiency.
Add an HDR enhancer if you are doing knock-in. Small molecules like L755507 or Brefeldin A have been reported to increase HDR efficiency by suppressing NHEJ, though the effect is cell-type dependent.
Synchronize cells in S phase for HDR. Treatment with 2 mM thymidine for 18 hours, followed by release, can enrich for cells in S phase when HDR is most active.
Lower the temperature to 30°C for 24 hours after transfection. Some reports indicate that Cas9 activity is more precise at lower temperatures, though this is not universally beneficial.

Validating Gene Editing: How to Confirm Your CRISPR Worked

You cannot assume that transfection equals editing. Validation is a mandatory step. The method you choose depends on whether you expect indels (knockout) or precise insertions (knock-in).

Detection of Indels

The most common and cost-effective method is the T7 Endonuclease I (T7E1) assay. The principle is simple: PCR-amplify the target region, denature and re-anneal the amplicons, and digest with T7E1, which cleaves mismatched heteroduplex DNA. If indels are present, the re-annealed mixture will contain mismatched duplexes, and digestion will produce smaller fragments.

Protocol outline:

Extract genomic DNA using a kit (e.g., DNeasy) or a simple phenol-chloroform method.
PCR-amplify a 300–600 bp region centered on the cut site. Use a high-fidelity polymerase (e.g., Phusion) to avoid introducing PCR-induced mismatches.
Purify the PCR product and quantify it.
Denature 200 ng of PCR product in 1× NEBuffer 2 in a total volume of 19 µL. Heat to 95°C for 5 minutes, then cool slowly to 25°C (ramp rate of −2°C per second) to allow heteroduplex formation.
Add 1 µL (10 U) of T7E1 enzyme and incubate at 37°C for 15–30 minutes.
Run the digestion on a 2% agarose gel. Undigested product will appear as a single band; digested samples will show additional smaller bands. Quantify the band intensities to estimate editing efficiency using the formula: % editing = 100 × (1 − √(1 − fraction cleaved)).

The Surveyor assay (Cel-1 nuclease) works on the same principle but uses a different enzyme. T7E1 is generally preferred because it is cheaper and has higher cleavage efficiency.

Sequencing and Genotyping

The T7E1 assay tells you that indels are present, but not their nature. For precise characterization, you need to sequence individual alleles.

Sanger sequencing of PCR products: If you sequence a PCR product from a pool of edited cells, the chromatogram will show overlapping peaks downstream of the cut site. You can use software like ICE (Inference of CRISPR Edits) from Synthego to deconvolute the mixed traces and estimate the frequency and types of indels.
TA cloning and Sanger sequencing: Clone the PCR product into a plasmid vector (e.g., pGEM-T), transform into E. coli, and sequence individual colonies. This gives you the exact sequence of each allele. It is laborious but definitive.
Next-generation sequencing (NGS): For high-throughput or clinical-grade validation, amplicon sequencing on an Illumina platform is the gold standard. It provides precise indel spectra and frequencies, but is overkill for most undergraduate experiments.

For knock-in experiments, you must also confirm that the donor sequence was inserted at the correct locus and not randomly integrated elsewhere. Use PCR with one primer outside the homology arm and one primer inside the inserted sequence. A product of the expected size confirms correct targeting. Sanger sequencing of this junction is recommended.

Applications of CRISPR in Research and Medicine

CRISPR has transformed molecular biology. The core applications fall into a few broad categories.

Gene Knockout and Knock-in

The most straightforward application is gene knockout via NHEJ-induced indels. This is used to study gene function, create disease models, and validate drug targets. For example, knocking out TP53 in a cell line can reveal its role in cell cycle arrest and apoptosis. The process is described in detail in CRISPR Knockout.

Knock-in via HDR allows you to introduce point mutations, insert fluorescent tags (e.g., GFP at the C-terminus of a protein), or create conditional alleles with loxP sites. The efficiency is typically 1–10% in mammalian cells, which is why selection or enrichment strategies are often necessary.

CRISPR Screens and Gene Regulation

Pooled CRISPR screens allow you to knock out thousands of genes in parallel and identify those that confer resistance or sensitivity to a drug. A library of sgRNAs targeting all genes in the genome is introduced into a pool of cells, and after selection, the sgRNA frequencies are quantified by NGS. Genes that are enriched or depleted reveal their functional relevance. This approach is covered in CRISPR Screening.

Beyond cutting, catalytically dead Cas9 (dCas9) can be fused to transcriptional activators (CRISPRa) or repressors (CRISPRi) to upregulate or downregulate gene expression without altering the underlying DNA sequence. dCas9-VP64 is a common activator; dCas9-KRAB is a common repressor. These tools are invaluable for studying gene regulation and for high-throughput functional screens. See CRISPR Knock for more on gene regulation applications.

CRISPR in Medicine

CRISPR is being developed for therapeutic applications, including the treatment of sickle cell disease, beta-thalassemia, and certain cancers. The first CRISPR-based therapy, exagamglogene autotemcel (Casgevy), was approved in 2023 for sickle cell disease and transfusion-dependent beta-thalassemia. It works by editing the BCL11A enhancer in hematopoietic stem cells to reactivate fetal hemoglobin expression. For a broader overview, see CRISPR in Medicine.

Common Pitfalls and How to Avoid Them

Even experienced labs encounter failures. Here are the most common problems and how to diagnose them.

Off-Target Effects

Off-target editing is the most serious concern. It occurs when the sgRNA binds to a genomic site with partial homology to the target. Even a few mismatches can be tolerated, especially if they are in the distal region of the spacer.

To minimize off-target effects:

Use a computational tool to select guides with high specificity scores.
Use a high-fidelity Cas9 variant such as SpCas9-HF1 or eSpCas9(1.1). These variants have mutations that reduce non-specific DNA contacts, lowering off-target activity while maintaining on-target efficiency.
Use the nickase pair strategy as described earlier.
Validate with off-target analysis: For the top predicted off-target sites, PCR-amplify and sequence them to confirm no editing occurred.

For a deeper dive, see CRISPR Off Target Effects.

Troubleshooting Low Editing Efficiency

If your T7E1 assay shows no editing, work through this checklist:

Confirm transfection efficiency: If fewer than 30% of cells are GFP-positive, your delivery is the bottleneck. Optimize transfection or switch to electroporation.
Check sgRNA expression: If using a plasmid with a U6 promoter, verify that your sgRNA does not contain a poly-T stretch (TTTT) that would terminate transcription.
Verify the target sequence: Re-sequence the genomic locus in your cell line. Cell lines accumulate mutations; the target site may differ from the reference genome.
Check for PAM: Confirm that your target site has an NGG PAM immediately downstream. A common error is designing a guide without a valid PAM.
Increase harvest time: Editing continues for 48–72 hours after transfection. Harvesting too early (e.g., at 24 hours) may give a false negative.
Use a positive control: Include a guide targeting a well-characterized locus (e.g., HPRT1 or AAVS1) to confirm that your reagents and protocol work.

Ethical Considerations and Future Directions

Ethical Debates

The use of CRISPR in human embryos, gametes, and germline cells raises profound ethical questions. Germline edits are heritable, meaning they would be passed to future generations. The scientific community largely agrees that germline editing is premature and unsafe, given the current risk of off-target effects and mosaicism. The 2018 case of the "CRISPR babies" in China, where twin girls were born with edited CCR5 genes, was widely condemned and led to calls for a global moratorium on heritable editing.

Somatic editing (in non-reproductive cells) is less controversial and is already being used in clinical trials. However, issues of equity, access, and informed consent remain. For a balanced discussion, see CRISPR Ethical Concerns.

Next-Generation CRISPR Tools

The field is evolving rapidly. Two notable advances are:

Base editing: A fusion of dCas9 or nickase with a deaminase enzyme (e.g., cytidine deaminase for C→T conversions, or adenine deaminase for A→G conversions). Base editors make single-nucleotide changes without creating a DSB, avoiding the need for HDR and reducing off-target effects. They are ideal for correcting point mutations that cause genetic diseases.

Prime editing: A fusion of nickase Cas9 with a reverse transcriptase and a prime editing guide RNA (pegRNA). The pegRNA contains both the target sequence and the desired edit. Prime editing can introduce insertions, deletions, and all 12 possible base substitutions without requiring a DSB or donor template. It is more versatile than base editing but currently less efficient.

These tools are not yet standard in undergraduate labs but are increasingly common in research settings. Understanding the basic CRISPR mechanism will prepare you for these advanced technologies.

Summary: Key Takeaways for Using CRISPR

CRISPR is a powerful and accessible tool, but it demands rigor. The difference between a successful and a failed experiment often comes down to careful design and validation. The essential steps are: choose your nuclease and delivery method, design a high-specificity sgRNA with a valid PAM, deliver the components efficiently, and validate the edit with a combination of enzymatic assays and sequencing.

Frequently Asked Questions

How do you use CRISPR for gene editing?

You use CRISPR by introducing a Cas9 nuclease and a guide RNA into cells. The guide RNA directs Cas9 to a specific genomic sequence adjacent to a PAM motif. Cas9 creates a double-strand break, which the cell repairs via NHEJ (producing indels and gene knockout) or HDR (producing precise edits if a donor template is provided).

What are the steps to use CRISPR?

The steps are: (1) design a guide RNA targeting your gene of interest, (2) choose a delivery method (plasmid, mRNA, or RNP), (3) introduce the components into cells via transfection or electroporation, (4) allow 48–72 hours for editing, (5) validate the edit using T7E1 assay, PCR, and sequencing.

How do you use CRISPR-Cas9 in the lab?

In the lab, you typically clone the sgRNA into a plasmid expressing Cas9, transfect the plasmid into your cell line, and select for edited cells. Alternatively, you can deliver Cas9 protein complexed with sgRNA (RNP) via electroporation. After 2–3 days, extract genomic DNA and assay for indels.

What is the PAM sequence and why is it important?

The PAM (protospacer adjacent motif) is a short DNA sequence required for Cas9 to bind and cleave. For SpCas9, it is 5′-NGG-3′. The PAM is essential because Cas9 cannot recognize the target site without it. It also prevents Cas9 from cleaving the CRISPR locus itself in bacteria.

How do you design a guide RNA for CRISPR?

Choose a 20-nucleotide sequence in your gene of interest that is immediately followed by an NGG PAM. Avoid poly-T stretches and target early, constitutive exons for knockout. Use online tools like Benchling or CRISPOR to score guides for efficiency and off-target potential.

How do you deliver CRISPR into cells?

Delivery methods include plasmid transfection (using lipid reagents), electroporation of mRNA or RNP complexes, and viral transduction (e.g., lentivirus or AAV). The choice depends on cell type: lipid transfection works for standard cell lines, while electroporation or viral delivery is needed for primary cells and in vivo work.

How do you know if CRISPR worked?

You confirm editing by PCR-amplifying the target region and running a T7E1 or Surveyor nuclease assay, which detects mismatched heteroduplexes. For precise characterization, perform Sanger sequencing of the PCR product or clone and sequence individual alleles. For knock-ins, use junction PCR with primers spanning the insertion site.

Key Takeaways

CRISPR-Cas9 requires three components: a Cas9 nuclease, a 20-nucleotide guide RNA, and an NGG PAM sequence adjacent to the target.
Guide RNA design is the most critical step; always use computational tools to maximize on-target efficiency and minimize off-target effects.
Choose your delivery method based on cell type: plasmids for standard lines, RNP electroporation for primary cells and stem cells.
NHEJ repair produces indels for gene knockout; HDR with a donor template enables precise knock-in edits.
Always validate editing with a T7E1 assay and sequencing; never assume transfection equals editing.
Off-target effects can be reduced using high-fidelity Cas9 variants, nickase pairs, or Cas12a.
CRISPR has broad applications in research, drug discovery, and medicine, but ethical considerations, especially around germline editing, remain critical.