How Bacteria Taught Us to Edit Genes

iStock / iStock

In 2005, Jennifer Doudna, a biochemist at the University of California, Berkeley, was looking at a bacterial genome recently sequenced by her colleague Jillian Banfield. Banfield was sequencing genomes of bacteria that lived in different environments, and she found an interesting peculiarity in one species—its genome contained repetitive DNA elements.

“At the time, no one knew what they were for, but several labs were looking at them,” Doudna tells mental_floss. Soon, scientific journals began publishing new findings. In between the repeated DNA segments were genetic sequences that bacteria apparently derived from viruses that infect them.

At the time, the detection of this phenomenon was seen as fundamental science research. Scientists named this interesting new system CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) and hypothesized that this genetic “archive” played a role in the bacteria’s immune defenses against viral infections.

Within a few years, the study of CRISPR had moved beyond fundamental research into a full-fledged gene-editing revolution that enabled scientists to fashion novel plants and animals with thrilling—and sometimes troubling—ease.

In labs around the world, scientists have used CRISPR to tweak genomes of mice, rats, and zebrafish. A company called Recombinetics produced a hornless cow with the idea that the animals would never suffer through the painful horn-cutting procedure. Biologists from two University of California schools (San Diego and Irvine) forged a mosquito with two genetic tweaks that let it fight off the malaria parasites so it can no longer spread them; that genetic trend is intended to propagate through the insect population. Meanwhile, Chinese scientists created dogs with more muscles, goats with more hair, and miniature pet pigs.


Humans learned these gene-editing techniques from bacterial species that used CRISPR to fight off their viral attackers. (Not all bacteria do.) Whenever such a bacterial cell kills off a virus, it inserts a fragment of the viral DNA into its own genome, which allows it to identify that virus easier in the future. To make that genomic self-edit, bacteria cut their own DNA using two CRISPR-associated proteins (Cas1 and Cas2), insert the virus’s genetic signature, and stitch the DNA back together with DNA-repairing enzymes.

John van der Oost, an early CRISPR researcher at the University of Wageningen, Netherlands, found that these genetic viral signatures serve as a memory of prior infection, or as vaccination against future viruses. Without these spacers, Escherichia coli bacteria, for instance, would succumb to a virus. With them, it can fight an infection off. Van der Oost tested this out. “When we gave an E. coli CRISPR spacers, it would gain immunity,” he says. “We called it a flu shot for the bacteria.”

The human immune system works in a somewhat similar way—albeit we’re much more complex than unicellular bacterial organisms. Yet our immune systems also have a way of identifying and remembering pathogens. That’s what makes vaccines work. A vaccine injects us with a weakened form of the pathogen, which our immune system fights off. After that, our immune system remembers how to kill this pathogen if it encounters it in real life—for example, how to make appropriate antibodies.

Likewise, bacteria actively use their “memorized” viral info to extinguish new invaders. They copy the DNA parts that contain the viral code into RNAs—the little mobile molecules that roam inside the cell checking for intruders, like seek-and-destroy missiles. “These RNAs are like a tape that doesn’t stick to just anything, but sticks to a matching genetic sequence,” Doudna says. If the RNA’s code signature matches the intruder's DNA, the latter will be destroyed.


Several CRISPR teams in the United States and Europe worked to understand how that seek-and-destroy process works. They found out that bacteria use a protein called Cas9 in combination with the RNA that carries the viral sequence info. When Cas9 encounters foreign DNA inside the bacterial cell, it physically unwinds that double-stranded DNA ribbon, and checks whether its genetic info matches what’s written in the RNA tape. If it does, Cas9 clips that foreign DNA in a manner similar to how scissors cut paper. In this process, the RNA essentially serves as a guiding force for Cas9, which is why it was dubbed a guide RNA. (While Cas1 and Cas2 cut and paste viral sequences from new viruses—ones the bacteria doesn’t have a “flu shot” for yet—Cas9’s job is to clip viral DNA every time a virus attacks.)

In this research, some pieces of the CRISPR-Cas9 puzzle came from Luciano Marraffini and Erik Sontheimer, at the time at Northwestern University in Illinois; some from Sylvain Moineau at University of Laval in Canada; and others from Doudna’s partnership with French researcher Emmanuelle Charpentier, who studied the deadly flesh-eating bacteria Streptococcus pyogenes. And as researchers pieced it all together, they ended up in a still-ongoing patent fight about who discovered what first.

Cas9 was not the first gene-editing technique scientists came across. There had been other ways to edit genomes—called TALENs or ZFNs—but they were much more cumbersome and hard to use. Doudna explains that these methods were essentially “hardwired,” requiring the researchers to create a new protein every time they wanted to make a single change to a genome. Cas9, on the other hand, was easily programmable. All one had to do was to change the guide RNA that Cas9 was coupled with, and the protein would aim at a different sequence on the foreign DNA ribbon and cut it at a different place.

“It was so trivial that many people started using Cas9 to experiment with organisms of interest,” Doudna says. That’s how we wound up with modified zebrafish, muscle-bound dogs, hairier goats, and micropigs.

The CRISPR-Cas9 technique was soon recognized as very promising in treating a gamut of genetic diseases—for example, muscular dystrophy or cystic fibrosis, in which certain genes fail to perform their normal functions. The theory is that we could use Cas9 to cut out a non-working genetic sequence and replace it with a working one. But scientists still have to figure out how to deliver the RNA and Cas9 editing complex into the specific cells in the body—into the affected muscles, for instance. Doudna is confident that eventually they will.


Gene editing also quickly raised a gamut of medical, legal, and ethical questions. The steady stream of studies in which scientists used CRISPR to change over a dozen plant and animal genomes, brought up an uncomfortable question: Are humans next? Would it be ethical and beneficial to apply gene-editing techniques to ourselves?

In December 2015, the major CRISPR players organized the International Summit on Human Gene Editing, which discussed the human gene-editing controversy and laid out several guidelines for basic research and clinical use. One takeaway from the summit is that altering genetic sequences in somatic cells—meaning cells whose genomes are not passed on to the next generation—does offer many benefits in curing diseases, and its outcomes can be systematically studied.

However, altering cells that can be passed on to future generations is a different story. It would be very difficult to systematically study outcomes of such actions, and any errors of genetic manipulation would be extremely hard to correct. So while gene editing can be used to eliminate heritable diseases as well as to enhance the human gene pool, it shouldn’t happen until proper scientific, societal, and legal guidelines are devised. Establishing such guidelines requires an ongoing conversation between scientists, policy-makers, and the public. Doudna says, “It’s not the decision that scientists can make alone."

Society will have plenty of time to battle over gene-editing dilemmas, because CRISPR research is far from over, Doudna says. Van der Oost is experimenting with a different protein, CPF1, which, he thinks, may one day rival Cas9, as it has similar properties. And there are other types of CRISPR systems that haven’t yet been studied, says Marraffini, now at Rockefeller University.

In a recently published paper, Marraffini described a CRISPR system that employs a delayed attack tactic. It doesn’t immediately destroy the identified viral DNA but waits to see whether the virus is beneficial; some may actually protect bacteria from other viruses.

“There may be other bacterial defense systems,” Marraffini says. “Whether they can be used for gene editing, we don’t know. But that’s why we need to study them.”