CELL BIOLOGY • APPLIED CELL BIOLOGY AND DISEASE

Gene Therapy & CRISPR — Explain gene therapy/CRISPR concepts at a high level (intro)

How molecular scissors and viral vectors are rewriting the genome to treat genetic disease.

Historical Context & Motivation

For much of the twentieth century, physicians treating genetic disorders could only manage symptoms—administering insulin for diabetes, transfusing blood for sickle-cell anemia, or prescribing enzyme supplements for metabolic deficiencies. The underlying mutations persisted in patients' genomes, passed faithfully from cell to daughter cell. The conceptual breakthrough of gene therapy was the realization that if disease originates in DNA, the most elegant cure would be to correct or replace the defective sequence itself. This ambition set the stage for decades of molecular-biology research that would ultimately converge on a bacterial immune system repurposed as the most precise genome-editing tool ever devised: CRISPR-Cas9.

1972
Gene Therapy Conceptualized
Theodore Friedmann and Richard Roblin publish "Gene Therapy for Human Genetic Disease?" in Science, arguing that functional genes could theoretically replace defective ones in patient cells, but cautioning against premature clinical trials.
1990
First Approved Gene-Therapy Trial
Four-year-old Ashanti DeSilva receives retroviral gene therapy for adenosine deaminase deficiency (ADA-SCID) at the NIH, marking the first federally approved human gene-therapy protocol.
1993
CRISPR Sequences Identified
Francisco Mojica catalogues repetitive DNA sequences in archaea, later named Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR). Their biological function remains mysterious for another decade.
2012
CRISPR-Cas9 as a Programmable Nuclease
Jennifer Doudna and Emmanuelle Charpentier demonstrate that the Cas9 protein, guided by a synthetic single guide RNA (sgRNA), can be directed to cut virtually any DNA sequence in vitro—ushering in the era of programmable genome editing.
2020
Nobel Prize & Clinical Milestones
Doudna and Charpentier receive the Nobel Prize in Chemistry. Meanwhile, the first in vivo CRISPR-based therapy is administered to a patient with Leber congenital amaurosis, and the FDA approves Casgevy for sickle-cell disease by 2023.

The trajectory from Friedmann's theoretical paper to FDA-approved CRISPR treatments spans roughly fifty years and encompasses triumphs, setbacks—including the tragic death of Jesse Gelsinger in a 1999 adenovirus trial—and paradigm shifts in delivery technology. The central question this lesson addresses is deceptively simple: How can we introduce, correct, or silence specific genes within living human cells to treat or cure disease?

Core Principles & Definitions

Gene therapy and genome editing rest on several foundational ideas drawn from molecular genetics, virology, and enzymology. Understanding these principles provides the conceptual scaffold upon which every clinical application is built. At the most general level, gene therapy seeks to alter the genetic instruction set inside a patient's cells; the distinction among strategies lies in how that alteration is achieved and whether the change is permanent or transient.

1

Gene Addition (Augmentation)

A functional copy of a gene is delivered into cells that carry a loss-of-function mutation. The defective allele remains, but the new copy supplies the missing protein. This strategy is typical of early viral-vector gene therapies such as Luxturna (for inherited retinal dystrophy).
2

Gene Editing (Correction / Disruption)

A programmable nuclease—such as Cas9—introduces a targeted double-strand break (DSB) at the mutated locus. The cell's own repair machinery then corrects the mutation via homology-directed repair (HDR) or disables the gene via non-homologous end joining (NHEJ).
3

Delivery Vectors

Therapeutic nucleic acids must cross the plasma membrane and reach the nucleus. Viral vectors (AAV, lentivirus) exploit natural cell-entry pathways, while non-viral vectors (lipid nanoparticles, electroporation) avoid immunogenicity concerns.
4

Ex Vivo vs. In Vivo

Ex vivo therapy modifies cells outside the body (e.g., harvesting hematopoietic stem cells, editing them, and reinfusing). In vivo therapy delivers the editing machinery directly into the patient's tissues, which is technically more challenging but eliminates the need for cell extraction.
5

Guide RNA Specificity

In CRISPR-Cas9, a ~20-nucleotide single guide RNA (sgRNA) base-pairs with a complementary DNA target adjacent to a protospacer adjacent motif (PAM). This Watson-Crick complementarity confers programmable sequence specificity, but imperfect matches can cause off-target editing.
KEY TAKEAWAY
Think of the genome as a massive instruction manual with billions of letters. Traditional gene therapy is like gluing a corrected page into the book—the typo on the original page is still there, but the cell can now read the right instructions. CRISPR, by contrast, is like a molecular word processor equipped with "find and replace": it locates the exact typo, cuts the page, and lets the cell's own proofreading machinery paste in the correct letters. The precision of the search function (the guide RNA) determines whether the edit lands on the right word or accidentally modifies a different paragraph entirely.

Visual Explanation — The CRISPR-Cas9 Mechanism

The diagram above traces the three key stages of CRISPR-Cas9 action. In step ①, the sgRNA hybridizes to the target strand adjacent to the PAM sequence. In step ②, the Cas9 endonuclease cleaves both DNA strands, producing a double-strand break (DSB). In step ③, the cell resolves the DSB through either NHEJ (often introducing disruptive indels) or HDR (enabling precise sequence correction when a donor template is supplied).

The beauty of the CRISPR system lies in its modularity. To retarget Cas9 to a different genomic locus, one simply synthesizes a new 20-nucleotide guide sequence—no protein engineering required. This stands in stark contrast to earlier genome-editing platforms such as zinc-finger nucleases (ZFNs) and TALENs, which require laborious protein redesign for each new target. The simplicity of guide-RNA design has democratized genome editing, enabling even modestly funded laboratories to perform gene knockouts, knock-ins, and regulatory perturbations with high efficiency.

Mechanistic Deep Dive — Delivery & Repair

Viral Vector Delivery

The most widely used viral vector for in vivo gene therapy is the adeno-associated virus (AAV). AAV is a small, non-enveloped parvovirus with a ~4.7 kb single-stranded DNA genome. For therapeutic use, the viral coding sequences (rep and cap) are removed and replaced with the transgene expression cassette flanked by inverted terminal repeats (ITRs). Because AAV genomes persist predominantly as extrachromosomal episomes in non-dividing cells, the risk of insertional mutagenesis is low relative to integrating vectors such as lentiviruses. However, AAV's limited packaging capacity (~4.7 kb) constrains the size of the therapeutic transgene, which has spurred development of dual-vector strategies and smaller Cas orthologs (e.g., SaCas9 at ~3.2 kb).

Non-Viral Delivery: Lipid Nanoparticles

An increasingly prominent non-viral strategy uses lipid nanoparticles (LNPs) to encapsulate mRNA encoding Cas9 along with the sgRNA. LNPs exploit endocytosis and endosomal escape to deliver their cargo to the cytoplasm, where the mRNA is translated by host ribosomes. Because the Cas9 mRNA is transient—degraded within hours to days—LNP delivery produces a pulse of editing activity followed by clearance, which may reduce off-target accumulation compared to constitutive expression from a viral vector. This approach was validated by Intellia Therapeutics in their NTLA-2001 program for transthyretin amyloidosis, representing a milestone in in vivo LNP-delivered CRISPR therapy.

DNA Repair Pathway Kinetics

After Cas9 generates a DSB, the ratio of NHEJ to HDR outcomes is a critical determinant of therapeutic success. In most mammalian somatic cells, NHEJ dominates because it is active throughout the cell cycle, whereas HDR is largely restricted to the S and G2 phases when a sister chromatid is available as a repair template. For applications requiring precise correction (e.g., inserting a normal allele), researchers have explored cell-cycle synchronization, small-molecule inhibition of NHEJ factors like DNA-PKcs, or alternative editing strategies such as base editing and prime editing, which avoid DSBs altogether.

EDITING EFFICIENCY
Editing Efficiency (%) = (Edited Alleles ÷ Total Alleles) × 100
Edited alleles are quantified by next-generation sequencing (NGS) or T7 endonuclease I assay. Clinically, efficiencies of 30–70% in target cell populations are often sufficient for therapeutic benefit, depending on the disease.
SPECIFICITY SCORE
Specificity = On-target edits ÷ (On-target edits + Off-target edits)
Values approaching 1.0 indicate high specificity. Guide-RNA design algorithms (e.g., CRISPOR, Benchling) score candidate guides by predicted off-target binding using mismatch tolerance models.

Delivery Vector Classification & Comparison

This flowchart contrasts the ex vivo and in vivo therapeutic pipelines. Ex vivo approaches involve cell harvest, editing, expansion with quality control, and reinfusion—typical for hematopoietic stem-cell therapies. In vivo approaches bypass cell extraction by delivering vector-packaged editing machinery directly to the target organ, as in AAV-based retinal gene therapies or LNP-mediated liver editing.
Comparison of common gene therapy delivery vectors
Vector TypeCargo CapacityIntegration RiskImmunogenicityClinical Example
AAV~4.7 kbVery low (episomal)Moderate; pre-existing antibodies commonLuxturna (RPE65 mutation)
Lentivirus~8 kbHigh (integrates)Low (pseudotyped VSV-G)Casgevy (ex vivo HSC editing)
LNPFlexible (mRNA/sgRNA)NoneLow; dose-dependent inflammationNTLA-2001 (TTR amyloidosis)
ElectroporationFlexible (RNP/plasmid)None (if RNP)None (physical method)CAR-T cell manufacturing

Worked Example — Designing a CRISPR Strategy for Sickle-Cell Disease

Sickle-cell disease (SCD) results from a single nucleotide substitution (A→T) in the HBB gene, converting codon 6 from GAG (glutamic acid) to GTG (valine). The mutant hemoglobin S (HbS) polymerizes under low-oxygen conditions, distorting red blood cells into a sickle shape. Let us walk through the reasoning a gene-therapy scientist would use to design a CRISPR-based therapeutic strategy.

CRISPR Strategy Design for SCD
1
Step 1 — Identify the Therapeutic GoalThe objective is to restore functional hemoglobin in the patient's red blood cells. Two strategies are available: (a) directly correct the HbS mutation to HbA using HDR, or (b) reactivate fetal hemoglobin (HbF) expression by disrupting the erythroid enhancer of the BCL11A repressor gene. The latter approach (used in Casgevy) requires only NHEJ-mediated disruption, avoiding the low efficiency of HDR.
Strategy: disrupt BCL11A erythroid enhancer via NHEJ to derepress HbF
2
Step 2 — Select the Target Sequence and PAMUsing a guide-RNA design tool (e.g., CRISPOR), we identify a 20-nt protospacer within the GATA1-binding site of the BCL11A erythroid enhancer on chromosome 2. The selected target must be immediately 5′ of an NGG PAM for SpCas9. Candidate guide: 5′-CUAACAGUUGCUUUUAUCAC-3′ (RNA), targeting the antisense strand.
sgRNA selected with high on-target score (>70) and low off-target matches
3
Step 3 — Choose the Delivery ModalityBecause hematopoietic stem cells (HSCs) can be harvested from the patient, an ex vivo approach is appropriate. We electroporate Cas9 ribonucleoprotein (RNP) complexes directly into CD34⁺ HSCs. RNP delivery provides transient Cas9 activity, minimizing off-target accumulation, and avoids the packaging-size constraints of AAV.
Delivery: ex vivo electroporation of Cas9 RNP into CD34⁺ HSCs
4
Step 4 — Assess Editing EfficiencyAfter electroporation, an aliquot of cells is harvested and the target locus is amplified by PCR and sequenced via NGS. Suppose we observe 450 edited alleles out of 600 total alleles sequenced. Editing efficiency = (450 ÷ 600) × 100 = 75%. This exceeds the ~30% threshold generally considered sufficient for meaningful HbF induction.
Editing efficiency = 75% — clinically actionable
5
Step 5 — Evaluate Outcome & SafetyEdited HSCs are expanded in culture, screened for off-target mutations at the top 10 predicted sites (no significant off-target editing detected), and reinfused into the patient following myeloablative conditioning. Post-transplant monitoring shows HbF levels rising to ~40% of total hemoglobin within 6 months, effectively preventing sickling crises.
HbF rises to ~40% — therapeutic success; patient achieves transfusion independence

Strengths, Limitations & Ethical Considerations

Strengths and limitations of gene therapy and CRISPR-based approaches
DimensionStrengthsLimitations / Risks
PrecisionCRISPR can target virtually any genomic locus with a 20-nt guide RNA; high specificity when guides are computationally optimizedOff-target DSBs may cause unintended mutations, chromosomal rearrangements, or activation of oncogenes
Therapeutic scopeApplicable to monogenic disorders (SCD, CF), certain cancers (CAR-T engineering), and infectious diseases (antiviral strategies)Polygenic and multifactorial diseases remain largely beyond current capability; delivery to all affected tissues is often impractical
DurabilityGenomic edits are permanent and inherited by daughter cells; a single treatment may suffice for a lifelong curePermanence is also a risk—erroneous edits cannot be "undone" once the cell divides
Cost & accessPotential for one-time curative treatments that eliminate chronic disease management costs over a lifetimeCurrent therapies priced at $1–3 million per patient; manufacturing complexity limits scalability; global equity concerns
EthicsSomatic gene therapy affects only the treated individual; broad scientific and regulatory consensus supports its developmentGermline editing (modifying eggs, sperm, or embryos) raises profound ethical debates; the He Jiankui case (2018) illustrated dangers of premature clinical application
THE GERMLINE BOUNDARY
A critical distinction in bioethics is between somatic and germline gene therapy. Somatic editing changes only the patient's own cells and is not heritable—analogous to patching a bug in a single copy of a software program running on one computer. Germline editing, by contrast, alters the master template from which all future copies will be made—akin to changing the source code in the repository. This is why the scientific community has largely endorsed a moratorium on human germline editing for reproductive purposes: the consequences propagate across generations and cannot be recalled.

Connection to Advanced Editing Technologies

While CRISPR-Cas9 remains the workhorse of genome editing, its reliance on double-strand breaks introduces risks of large deletions, chromosomal translocations, and p53-mediated cell-cycle arrest. These limitations have driven the development of next-generation editing tools that either avoid DSBs entirely or expand the repertoire of possible modifications. Understanding where standard CRISPR fits within this evolving landscape is essential for appreciating both its current clinical value and its future trajectory.

Comparison of CRISPR-Cas9 with next-generation editing platforms
FeatureCRISPR-Cas9 (Standard)Base EditingPrime Editing
MechanismDSB followed by NHEJ or HDRCatalytically impaired Cas9 (nickase) fused to a deaminase; converts C→T or A→G without DSBCas9 nickase fused to reverse transcriptase; uses a prime editing guide RNA (pegRNA) to write new sequences directly
Edit typesKnockouts, large insertions/deletions, knock-insAll 4 transition mutations (C→T, G→A, A→G, T→C)All 12 point mutations, small insertions (≤44 bp), small deletions (≤80 bp)
DSB required?YesNo (single-strand nick)No (single-strand nick)
Off-target profileModerate; DSB-mediated risksLower genotoxicity; bystander edits in editing windowLowest among current platforms; minimal bystander edits
Clinical stage (2024)FDA-approved (Casgevy)Phase I/II trials (e.g., Verve PCSK9 editing)Preclinical; first IND applications anticipated

The field is moving rapidly toward DSB-free editing as the default paradigm, with base editors and prime editors likely to supplant standard Cas9 for many therapeutic indications. Additionally, emerging tools such as epigenome editors (CRISPRi/CRISPRa) offer reversible gene silencing or activation without altering the DNA sequence at all—adding yet another layer of therapeutic versatility. Students entering the field today should expect the molecular toolkit to continue expanding at an accelerating pace.

Practice Problems

PROBLEM 1CONCEPTUAL
Explain the fundamental difference between gene addition (augmentation) therapy and gene editing. In which clinical scenario would gene addition be preferred over CRISPR-mediated editing, and why?
PROBLEM 2BASIC CALCULATION
A research team electroporates Cas9 RNP into a population of CD34⁺ cells and sequences the target locus by NGS. They detect 1,260 alleles with indels at the target site out of 1,800 total alleles sequenced. Calculate the editing efficiency. If a minimum of 40% editing is required for therapeutic benefit, does this experiment meet the threshold?
PROBLEM 3INTERMEDIATE
A scientist wants to correct a G→A point mutation in a patient's hepatocytes in vivo. She considers three approaches: (a) standard Cas9 with an HDR template delivered by AAV, (b) an adenine base editor (ABE) delivered by LNP, or (c) a prime editor delivered by dual-AAV. Discuss the trade-offs of each approach with respect to efficiency, off-target risk, and practical constraints.
PROBLEM 4APPLIED
A biotech company is developing an in vivo CRISPR therapy for Duchenne muscular dystrophy (DMD), caused by frame-shifting deletions in the dystrophin gene. The full dystrophin cDNA is ~14 kb. The company proposes using AAV9 to deliver Cas9 and two sgRNAs to excise exon 51, restoring the reading frame to produce a truncated but partially functional dystrophin (analogous to the milder Becker phenotype). Identify at least three major scientific or regulatory challenges this program would face.
PROBLEM 5CRITICAL THINKING
In 2018, He Jiankui used CRISPR to edit the CCR5 gene in human embryos, aiming to confer HIV resistance. The resulting children (Lulu and Nana) carried mosaic edits—some cells were edited, others were not, and some harbored unintended indels. Construct a multi-layered argument addressing: (a) why mosaicism undermines the stated therapeutic goal, (b) the biological risks of biallelic CCR5 disruption, and (c) the broader implications for governance of germline genome editing.

Lesson Summary

Gene therapy encompasses strategies that introduce, replace, or modify genetic material within a patient's cells to treat disease. The field originated in the 1970s with theoretical proposals, endured setbacks in the late 1990s, and has now achieved FDA-approved clinical therapies. Two broad paradigms exist: gene addition, which supplies a functional transgene via viral vectors (AAV, lentivirus) or non-viral delivery (LNPs, electroporation), and gene editing, which uses programmable nucleases to alter the genome at specific loci.

CRISPR-Cas9 revolutionized genome editing by pairing a single guide RNA (sgRNA) with the Cas9 endonuclease to generate targeted double-strand breaks, which cells repair via NHEJ (gene disruption) or HDR (precise correction). Clinical milestones include Casgevy for sickle-cell disease and LNP-delivered in vivo editing for transthyretin amyloidosis. Next-generation tools—base editors and prime editors—avoid DSBs entirely, promising greater precision and safety. Ethical governance remains paramount, particularly regarding the bright line between somatic editing (permissible) and germline editing (moratorium).

Varsity Tutors • Cell Biology • Gene Therapy & CRISPR