Wikipedia Deep Dive

CRISPR gene editing

14 min read

In November 2018, a Chinese scientist named He Jiankui announced that he had done something no one had ever done before: he had edited the genes of human embryos that were then carried to term. Twin girls, known by the pseudonyms Lulu and Nana, became the first genetically modified humans ever born. The scientific community was horrified. He was sentenced to three years in prison. But the genie was out of the bottle.

The tool that made this possible—and that has sparked both tremendous hope and existential dread about humanity's future—is called CRISPR, pronounced like the vegetable drawer in your refrigerator. The name stands for "clustered regularly interspaced short palindromic repeats," which tells you almost nothing useful about what it actually does. What it does is this: it lets scientists edit genes with unprecedented precision, speed, and cheapness.

Think of it as a word processor for DNA.

The Accidental Discovery

CRISPR wasn't invented. It was discovered. And not by humans looking for a gene-editing tool, but by bacteria defending themselves against viruses.

Bacteria have been fighting viral infections for billions of years. Some of them evolved a clever defense mechanism: when a virus attacks and the bacterium survives, it stores a snippet of the virus's genetic code in its own DNA. These stored snippets are the "clustered regularly interspaced short palindromic repeats"—essentially a bacterial memory bank of past infections. If that same virus ever attacks again, the bacterium can recognize it and deploy a molecular weapon to destroy it.

That weapon is a protein called Cas9.

Cas9 works like molecular scissors. The bacterium produces a small piece of RNA—a molecule similar to DNA—that matches the stored viral sequence. This "guide RNA" attaches to Cas9 and tells it exactly where to cut. When Cas9 finds DNA that matches its guide, it slices through both strands of the double helix. In nature, this destroys the invading virus.

Scientists realized they could hijack this system. Instead of using guide RNA that targets viruses, they could design guide RNA that targets any gene they wanted. Give Cas9 the right instructions, and it becomes a precision cutting tool for any genome on Earth.

How the Scissors Work

To understand CRISPR gene editing, you need to understand what happens when DNA gets cut.

DNA is remarkably good at repairing itself. Cells have multiple repair mechanisms because breaks in DNA happen constantly—from radiation, chemical damage, or just the normal wear and tear of cellular life. When Cas9 makes a cut, the cell's repair machinery kicks in immediately.

There are two main ways cells repair a double-strand break, and scientists exploit both of them.

The first is called non-homologous end joining, which is exactly what it sounds like: the cell grabs the two broken ends and sticks them back together. This process is fast but sloppy. It often deletes a few base pairs or inserts random ones. If this happens in the middle of a gene, it typically breaks that gene—it can no longer produce a functional protein. Scientists call this a "knockout," and it's incredibly useful for studying what genes actually do. Want to know what happens when a mouse can't produce a certain protein? Knock out the gene and watch.

The second repair pathway is more sophisticated. It's called homology-directed repair, and it uses a template to guide the repair process. In nature, this template is usually the matching chromosome—humans have two copies of most genes, one from each parent—but scientists can provide their own template. If you supply a piece of DNA with the sequence you want, the cell's repair machinery will sometimes incorporate it into the genome. This lets you not just break genes, but rewrite them.

You can fix a mutation that causes disease. You can insert an entirely new gene. You can make precise, single-letter changes to the genetic code.

Before CRISPR: A Brief History of Gene Editing

CRISPR wasn't the first gene-editing technology. Scientists had been modifying genomes since the 1980s, but the older methods were slow, expensive, and often unreliable.

The first serious contenders were zinc finger nucleases, developed in the early 2000s by German researchers. These are synthetic proteins engineered to recognize specific DNA sequences and cut them. They work, but they're difficult to design. Each zinc finger recognizes three base pairs of DNA, so cutting a specific twenty-base-pair sequence requires engineering multiple zinc fingers and linking them together. This takes months of work and often fails.

In 2010, a newer technology called TALENs—transcription activator-like effector nucleases—made the design process somewhat easier. But both zinc finger nucleases and TALENs share the same fundamental limitation: each new target requires building a new protein from scratch. This is slow, expensive, and requires specialized expertise.

CRISPR changed everything because the targeting mechanism is completely different. Instead of engineering a protein to recognize each new DNA sequence, you just synthesize a short piece of RNA. Making custom RNA is trivial—labs do it constantly for all kinds of experiments. The same Cas9 protein works for any target. You just swap out the guide RNA.

A zinc finger nuclease project might take a year and cost hundreds of thousands of dollars. A CRISPR experiment can be designed in minutes and costs a few hundred dollars. The barriers to entry collapsed almost overnight.

The Nobel Prize and the Patent War

The key breakthrough came in 2012, when Jennifer Doudna at the University of California, Berkeley, and Emmanuelle Charpentier, then at Umeå University in Sweden, published a paper showing that CRISPR-Cas9 could be programmed to cut any DNA sequence. They figured out how to combine the two natural RNA molecules that guide Cas9—called crRNA and tracrRNA—into a single "guide RNA" that's easy to synthesize. This paper is now considered one of the most significant discoveries in the history of biology.

In 2020, Doudna and Charpentier won the Nobel Prize in Chemistry for this work. They made history as the first two women to share a science Nobel without a male co-recipient.

But the Nobel committee's decision papered over a bitter controversy.

Doudna and Charpentier demonstrated that CRISPR-Cas9 could cut DNA in a test tube. But the really valuable application—editing genes inside living cells—was first achieved by Feng Zhang at the Broad Institute, a research center affiliated with MIT and Harvard. Zhang published his results in January 2013, just months after Doudna and Charpentier's paper, and he moved quickly to patent the technique.

What followed was one of the most contentious patent disputes in the history of biotechnology. The University of California argued that its researchers conceived of the invention first. The Broad Institute argued that conceiving an idea isn't enough—you have to actually make it work, and Zhang was the first to do that in living cells. In 2017, the US Patent Office sided with the Broad Institute. The University of California appealed, lost again in 2022, and announced plans to appeal further.

The stakes are enormous. Whoever holds the foundational CRISPR patents controls licensing rights to a technology that could be worth billions of dollars.

Notably, a third research group led by Virginijus Šikšnys at Vilnius University in Lithuania made similar discoveries around the same time. Šikšnys shared the Kavli Prize with Doudna and Charpentier but was not included in the Nobel. The history of scientific discovery is rarely as clean as prize committees make it seem.

What CRISPR Can Do Today

The first CRISPR-based medicine was approved in November 2023, when the United Kingdom's Medicines and Healthcare products Regulatory Agency authorized a treatment called Casgevy for sickle cell disease and beta thalassemia.

Both diseases are caused by mutations in the genes that produce hemoglobin, the protein in red blood cells that carries oxygen. In sickle cell disease, a single-letter change in the genetic code causes hemoglobin molecules to stick together, deforming red blood cells into a crescent or "sickle" shape. These misshapen cells clog blood vessels, causing excruciating pain, organ damage, and shortened lifespans. Beta thalassemia involves different mutations that reduce hemoglobin production, leading to severe anemia.

Casgevy works by editing the patient's own stem cells—the cells in bone marrow that produce all blood cells. Doctors extract these stem cells, use CRISPR to modify them, and infuse them back into the patient. The edit doesn't fix the mutated hemoglobin gene directly. Instead, it reactivates a different hemoglobin gene that normally shuts off after infancy. This fetal hemoglobin works just as well for carrying oxygen, bypassing the defective adult version.

The treatment is not simple. Patients must undergo chemotherapy to destroy their existing bone marrow before receiving the edited cells. The process takes months and costs over a million dollars. But for patients who have suffered their entire lives, who have endured countless blood transfusions and hospitalizations, who face the prospect of dying young—it offers something that was previously impossible: a cure.

Bahrain became the second country to approve Casgevy in December 2023, followed by the United States just days later. In February 2025, Bahrain announced the first successful treatment of a sickle cell patient outside the United States, with the World Health Organization's director general offering congratulations.

Beyond Medicine: CRISPR in Agriculture

The first CRISPR-edited food went on sale in Japan in September 2021. It was a tomato.

These tomatoes were engineered to contain about five times the normal amount of a compound called GABA—gamma-aminobutyric acid—which some studies suggest has calming effects on the nervous system. The company that developed them, Sanatech Seed, used CRISPR to disable a gene that normally limits GABA production.

Just months later, Japan approved CRISPR-edited fish. One is a sea bream modified to grow to twice its normal size by disrupting the gene for leptin, a hormone that regulates appetite. The fish eat more and grow bigger. Another is a tiger puffer modified to reach 1.2 times normal size with the same amount of food, achieved by disabling myostatin, a protein that inhibits muscle growth.

These modifications would have been possible with traditional genetic engineering, but CRISPR makes them faster, cheaper, and more precise. And because CRISPR edits can simply disable existing genes rather than inserting foreign DNA, they occupy a regulatory gray zone. In Japan and some other countries, CRISPR-edited foods that don't contain genes from other species face lighter regulation than traditional genetically modified organisms.

Not everywhere, though. In 2018, the European Court of Justice ruled that gene-edited crops should be regulated under the same strict rules as traditional GMOs. This has effectively blocked most CRISPR agriculture in Europe, to the frustration of many scientists who argue that the technology is fundamentally different from older genetic engineering methods.

The Germline Problem

There's a crucial distinction in gene editing: somatic versus germline.

Somatic edits affect only the person being treated. When doctors modify a sickle cell patient's bone marrow stem cells, those changes stay in that patient. They won't be passed to any children the patient might have.

Germline edits are different. If you edit an embryo, or egg cells, or sperm, those changes will be inherited by all future generations. This is what He Jiankui did in China—and why the scientific community reacted with such alarm.

The distinction matters for several reasons.

First, there's the question of consent. A person being treated for sickle cell disease can agree to experimental therapy. An embryo cannot consent to modifications that will affect every cell in its body and potentially be passed to descendants indefinitely.

Second, there's the question of safety. CRISPR is precise, but it's not perfect. Sometimes Cas9 cuts in the wrong place—these are called "off-target effects." In a somatic treatment, off-target edits might cause problems in some cells, but they're limited to one individual. In a germline edit, any mistake becomes permanent, propagating through the gene pool forever.

Third, there's the specter of eugenics. The same technology that could eliminate genetic diseases could theoretically be used to select for traits like intelligence, height, or appearance. The history of eugenics in the twentieth century—including forced sterilization programs in the United States and the horrors of Nazi Germany—casts a long shadow over any technology that could be used to "improve" the human species.

Most scientists support a moratorium on human germline editing until the technology is better understood and society has had time to debate the ethical implications. But the technology exists. It's relatively simple to use. And as He Jiankui demonstrated, someone willing to ignore ethical norms can use it.

The Technical Details: For Those Who Want Them

The standard CRISPR-Cas9 system has several components that work together.

The Cas9 protein is the cutting enzyme. The version most commonly used comes from a bacterium called Streptococcus pyogenes—the same species that causes strep throat. The protein is large, about 1,400 amino acids, and has two cutting domains that each slice one strand of the DNA double helix.

The guide RNA has two parts. The first twenty nucleotides at one end are the targeting sequence—this is what you customize for each experiment. The rest of the guide RNA forms a specific structure that Cas9 recognizes and binds to. In nature, these are two separate RNA molecules called crRNA and tracrRNA, but scientists usually fuse them into a single "single-guide RNA" or sgRNA for convenience.

There's one more constraint. Cas9 doesn't just cut anywhere its guide RNA matches. It also requires a specific short sequence right next to the target site, called a protospacer adjacent motif or PAM. For the Streptococcus pyogenes Cas9, this sequence is NGG—any nucleotide followed by two guanines. This means you can only target sequences that happen to have this motif nearby.

This might seem like a serious limitation, but in practice it's not. The NGG sequence is common enough that there's usually a suitable target site within a few dozen base pairs of wherever you want to edit. And researchers have developed alternative Cas proteins from other bacteria that recognize different PAM sequences, expanding the range of targetable sites.

Newer variants of Cas9 have also been engineered to reduce off-target cutting. The original enzyme sometimes cuts sequences that are similar but not identical to the intended target. Modified versions trade some cutting efficiency for much higher specificity, making unintended edits far less likely.

Beyond Cas9: The Expanding Toolkit

Cas9 was just the beginning. Scientists have discovered and engineered many other CRISPR-associated proteins with different capabilities.

Cas12a, originally called Cpf1, uses a different mechanism than Cas9 and recognizes a different PAM sequence. This expands the range of sequences that can be targeted.

Cas13 targets RNA instead of DNA. This opens up entirely different applications—you could use it to destroy viral RNA in infected cells, for example, without making any permanent changes to the genome.

Perhaps most exciting are the "base editors" and "prime editors" that don't cut DNA at all. Traditional CRISPR makes a double-strand break and relies on the cell's repair machinery to make changes. This process is inherently imprecise. Base editors, developed by David Liu's lab at Harvard, chemically convert one DNA base to another without breaking the strand. Prime editors go further, allowing precise insertion, deletion, or replacement of sequences up to several dozen bases long.

These newer tools are addressing the biggest limitations of original CRISPR. They're more precise, cause fewer unintended effects, and can make changes that were previously difficult or impossible.

What Comes Next

Clinical trials are underway for CRISPR treatments targeting cancer, HIV infection, genetic blindness, and many other conditions.

In February 2020, researchers reported the first results from a trial using CRISPR to edit immune cells from cancer patients. The study was small—only three patients—but it demonstrated that the approach was safe, at least in the short term. The edited cells survived in patients' bodies for months, proving that CRISPR-modified cells could persist and function.

In June 2021, the first trial of intravenous CRISPR delivery showed promising results. Previous treatments had required removing cells from the body, editing them in a lab, and putting them back. Direct injection would be far simpler and cheaper, potentially bringing CRISPR therapies to millions more patients.

The technology continues to improve. Each year brings more precise editing tools, better delivery methods, and deeper understanding of how to predict and prevent off-target effects. The first generation of CRISPR medicines is just arriving. Many more will follow.

But the technology that enables cures also enables enhancement. The same tools that could eliminate genetic diseases could theoretically modify any human trait with a genetic component. How humanity navigates this power—what we allow, what we prohibit, and who gets to decide—may be one of the defining questions of this century.

The molecular scissors have been invented. What we cut with them is up to us.